Full text of "Rhythm"
Early Journal Content on JSTOR, Free to Anyone in the World
This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in
the world by JSTOR.
Known as the Early Journal Content, this set of works include research articles, news, letters, and other
writings published in more than 200 of the oldest leading academic journals. The works date from the
mid-seventeenth to the early twentieth centuries.
We encourage people to read and share the Early Journal Content openly and to tell others that this
resource exists. People may post this content online or redistribute in any way for non-commercial
purposes.
Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early-
journal-content .
JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people
discover, use, and build upon a wide range of content through a powerful research and teaching
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please
contact support@jstor.org.
THE AMEKICAN
Journal of Psychology
Edited by G. Stanley Hall.
Vol. VI. JANUARY, 1894. No. 2.
BHYTHM.
By Thaddeus L. Bolton,
Demonstrator and Fellow in Psychology, Clark University.
Introduction.
Although experimental psychology began more than a
hundred years ago through the discovery of the personal
equation, it has as yet covered but a small portion of the field
of mental phenomena. The nature of sensation, the time-
relations of mental phenomena, memory, association, space
and time concepts have been carefully studied by many
eminent scientists. But the whole field of the emotions has
been practically a mare clausum for psychologists. Several
attempts to study and determine the nature of the simplest
aesthetic forms have been reported ; theories of pleasure and
pain, supported by some experimental observations, have been
advanced, but no serious attempt has been made to submit
the emotions to experimental investigation. Every psycholo-
gist recognizes the necessity of doing so and that it is the
most important field of mental phenomena, and that, until
psychologists can reduce the motions to some semblance of
order or more ultimate principles, experimental psychology
can be said to cover only a part of the field of mental life.
Every one is waiting for some one else to point out the way.
There seems to be a general feeling that when once an en
trance has been effected, the greatest difficulty will have been
146 BOLTON :
surmounted and the whole field will yield to experiment. In
a measure this is the true view to take, and yet certain of the
emotions are as distinctly separated from others as the whole
field is from that of the memory.
When the demand for such a study is so great, and students
are being urged to make a trial, that one who does attempt it,
though he fail, cannot be accounted rash.
The experimental study of rhythm which is to be presented
in this paper, is an attempt to push the lines of exact science
a little farther forward into a field that borders more closely
upon the field of aesthetics than any other that experimental
psychologists have tried. The attempt is to be made to re-
duce rhythm to a more fundamental activity of mind. The
pleasure that individuals take in the rhythmic flow of words
and sounds has been ascribed by one to the u Unifying Ac-
tivity of the Feelings, " by another to a " Sense of Order, "
and by still another to " The Feelings of Equality.' ? Such
explanations as these do not meet the question at all, unless
it can be shown that such activities or feelings are ultimate
facts of mind. If they are ultimate facts of the mind, it will
be necessary, in order to make the explanation complete and
valid, to show how they underlie other activities, for it is not
to be supposed that any fundamental activity will manifest
itself in a single phenomenon which bears no relation to
other phenomena. Such does not seem to have been done by
those who have offered explanations of the rhythms in speech,
and the problem remains just where it was taken up. To re-
gard rhythm as the manifestation or the form of the most
fundamental activities of mind, seems a clearer view and to
offer less difficulties than to regard it as an ultimate fact in
itself. The problem, then, is to show how and to what extent
it underlies mental activity, and, as preparatory to this, what
part it plays in physiology and nature. Is there not some
universal principle which is adequate as an explanation of
rhythm in general?
Ehythm is so universal a phenomenon in nature and in
physiological activity, and underlies so completely speech,
that I desire to call attention to some of its manifestations in
detail before presenting the experimental study.
Rhythms in Nature : — Natural phenomena very generally,
if not universally, take a rhythmic form. There is aperiodic
recurrence of a certain phenomenon, sometimes accompanied
by others, going on continuously in all that pertains to
nature. Motion, whether in the broader field of the universe
or upon the earth, is very generally periodic. Light, heat,
sound, and probably electricity, are propagated in the form of
waves. A falling body does not follow a straight line,
EHYTHM. 147
neither does a rifle bullet describe a simple curve which is the
resultant of the combined forces of gravity and the initial
velocity. Mr. Herbert Spencer has treated this subject in
his " Principles of Philosophy" at considerable length, and
has left but little that can be said here. Although he does
not say so in so many words, he seems to hold that it is the
only possible form of activity ; continuous motion is an im-
possibility.
The cosmic rhythms, however, are the most fundamental and
important of natural phenomena. They may be shown to
underlie in a measure and be the cause of many other rhythms
in plant and animal life. The regular alternation of light and
darkness due to the rotation of the earth upon its axis is the
most striking rhythm in the cosmos. The two periods of
light and darkness constitute a unit — the day — which re-
mains always the same in length. Days are grouped into
months by the revolution of the moon about the earth, and
into years by the revolution of the earth about the sun.
These periodic changes have had a tremendous influence
upon animal and plant life, and have stamped their impress
upon all living organisms in the most striking manner ;
some, however, upon certain organisms more than upon others.
In the vegetable kingdom some plants show a daily growth
and repose; their flowers bloom in the morning and close
before the evening. Some turn their petals towards the sun,
and make a daily revolution in order to keep them so. In
certain latitudes all vegetation shows normal periods of
growth and fruitage which are not necessarily cut short or
lengthened by early or late frosts. It requires a certain time
for development without regard to the character of the season.
The lunar period is known to influence the blooming of
flowers. A species of Chinese roses blooms with a monthly
regularity during the season/)
The influence of these cosmic rhythms is not less upon the
animal kingdom. The daily rhythm causes the daily periods
of sleep and waking, from which no terrestrial creatures of
the higher types are exempt. The periods of sleep and wak-
ing are not determined by the effect of light and darkness as
are the movements of many plants. The lunar period has had
a far-reaching effect upon animal creatures, especially as re-
gards reproduction and the nervous system. The periods of
gestation and the recurrence of heat and menstrual flow in
both human beings and animals bear a very close and strik-
ing relation to the lunar period. The period of gestation in
some lower mammalian animals is one month. In the higher
forms it is a certain number of months. The time of incuba-
tion is with some species of fowls a month, but it seems to
148 BOLTON :
conform in general to a period of days which is a certain
multiple of seven, seven being one-fourth of a lunar month.
Fourteen, twenty-one and twenty-eight days are very common
periods of incubation. The year exercises a still wider in-
fluence upon the animal kingdom. The normal life of most
species of insects terminates in a single year. The frog
becomes nervous and irritable with the approach of spring,
although the conditions under which it is kept may not
change. The polar bear goes into hibernation, even though
he has not made the proper preparation in the way of a store
of fat. The migrations of birds are not necessarily prompted
by the signs of approaching winter. Animals breed generally
in the spring — a fact which cannot find sufficient explanation
in the influence of a warmer temperature. It has been fairly
established that growth is more rapid during the summer
months.
Although we find that these cosmic rhythms have stamped
themselves upon the organism more or less permanently,
they have wielded a far mightier influence upon the minds of
men. Among primitive peoples that were rich in imagina-
tive power, they have given rise to the most elaborate and
beautiful systems of mythology and worship that the world
has ever seen. It is a common speculation in childhood
that, endowing animals at birth, as children do, with rational
intelligence, but with a total lack of experience, the young
creature must be driven to strange thoughts and speculations
when the first light of day breaks in upon him, or when dark-
ness approaches for the first time. What can be the thoughts
of such a creature when he experiences the change of seasons
or the first snow storm? No objects that are presented to
the child so stimulate his thought and become such food for
his fancy as the heavenly bodies and cosmic phenomena.
Many of their minds are filled with myths about the stars
that are as original and beautiful in conception, though lack-
ing in detail, as much of the Greek mythology.
The recurrence of the day of the year upon which some
event has happened is commemorated as a day of joy or
sorrow according to the nature of the event. All national
and religious festivals recur once a year. Among primitive
peoples worship takes place always at the same time of day
or year, and the same might be said of most enlightened
people. There seems now, and always has seemed, a
peculiar appropriateness in performing certain duties at the
same time of day or year, although it does not necessarily
depend upon the nature of the weather or of the event. The
Christian Sabbath and other religious festivals, both savage
and civilized, find their origin in the nature worship of the
sun and the moon.
RHYTHM. 149
There are still other rhythms in the cosmos which seem to
exercise an influence upon mankind. Sun spots make their
appearance in great numbers once in about eleven years, and
the attempt has been made to connect these with great
financial disasters and religious awakenings which seem to
recur in the same time. The social customs of the race
show similar changes, which may prove to have some con-
nection with sun spots. The coincidence warrants an in-
vestigation and allows speculation.
Upon the morbid side science has made discoveries of the
most striking character. Even from the earliest times a
periodicity has been observed in certain forms of insanity
and in other mental diseases. These have been confirmed by
later investigations. 1 Both crime and suicide show a
periodicity which corresponds with the year, and another
which corresponds to the larger period of sun spots.
Physiological Rhythms : — No fact is more familiar to the
physiologist than the rhythmic character of many physio-
logical processes. In physiology it means the regular alter-
nation of periods of activity and periods of repose or of lesser
activity. The term is also applied to any alternation of
activity and repose, whether it is regular or not. These
periods of activity and intervals of repose may succeed one
another at very small intervals of time, as in the case of a
clonic contraction of the muscle, or at very much greater
intervals, as in the case of sleep and waking, or better still, in
the periods of growth in children. Several of the most vital
and important bodily activities are distinctly rhythmical, and
will serve as types of all physiological rhythms. Of these,
might be mentioned the pulse, respiration, walking and
speech. The first two are involuntary actions, which in the
very nature of the organism must be more or less rhythmical.
Such actions are controlled by the lower nerve centres, and
the organs concerned in them are connected in a reflex arc
with these nerve centres. Habits are in the nature of invol-
untary actions. Of these, walking and speech are the most
important and are true types of rhythmical activity. In each
there is a series of coordinated muscles in which the contrac-
tion of one is the signal for the contraction of the next in the
series, the last acting as a stimulus to the first.
Independent of the regular beat of the heart and forming a
kind of higher grouping of these beats, the arteries undergo
continuously rhythmical contractions and dilations of their
1 Dr. Koster, "Uber die Gesetze des periodischen Irreseins und
verwandter Nervenzustande." Bohn, 1882.
Dr. Ludwig, " Periodischen Psychosen." Stuttgart, 1878.
150 BOLTON :
walls, now increasing and now decreasing the blood supply.
These may be observed with a glass in the arteries of a frog's
foot or a rabbit's ear, occurring about once a minute. They
may be made to cease entirely by cutting off the nerves going
to these organs. 1 These arteries are controlled by the vaso-
motor system, and the rhythmic contractions of the arteries
seem to indicate a rhythm in the activity of the nerve
centres. As we shall see later, there is some ground for
believing that all nervous action is rhythmical. Eegular
contractions occur in the heart of some animals after they
have been removed from the body, and are found to be due
probably to the presence of nerve ganglia in these organs. 2
The effect of deficient arterialization upon the vaso-motor
system is to cause a rise in the curve of blood pressure.
This curve, then, shows certain undulations, which have
been called Traube-Hering curves, from their discoverer.
This result is obtained by cutting the vagi nerve and
stopping respiration. The venus blood then acts as a
stimulus upon the vaso-motor centres in the medulla,
which causes these rhythmic movements. "This rhythmic
rise 3 must be due to the rhythmic contraction of the arteries,
and this is caused by a rhythmic discharge from the vaso-
motor centres. " u The vaso-motor nervous system is apt to
fall into a condition of rhythmic activity." A similar
phenomenon has been thought to be observed in regard to
the spinal cord.
When the spinal cord 4 of a dog, cat or rabbit was cut,
rhythmical contractions of the sphincter ani and of the
vagina appeared. These contractions vary in number, but
are generally about twenty per minute for the sphincter ani
and four per minute for the vagina. The centre for these
contractions was found to be in the spinal cord, about the
level of the sixth and seventh lumbar vertebrae in rabbits and
of the fifth lumbar vertebra in dogs.
1 Foster's "Physiology," 6th Ed. p. 307.
Dr. Ellis, working under Dr. Bowditch, has studied these contrac-
tions in the web of a frog's foot with the microscope. He says that
cutting the sciatic nerve does not stop them, and concludes that
they are due to peripheral centres, unless he be allowed to suppose
that automatic contractility is a property of smooth muscle tissue.
Plethysmographic and vaso-motor experiments with frogs. Jour,
of Phys. Vol. VI. No. 6, p. 437.
2 Foster's "Physiology," 6th Ed. p. 357.
3 Foster's "Physiology," 6th Ed. p. 622.
4 Isaac Ott, "Observations upon the Physiology of the Spinal
Cord." Studies from Biol. Lab. at Johns Hopkins University.
No. II.
KHYTHM. 151
Fatigue shows itself to be a rhythmical process. Dr.
Lombard 1 worked upon the flexor muscle of the second
finger. After contracting the muscle several times, lifting
each time a weight, he gradually lost the power of further
contraction, but he continued to make the effort at regular
intervals of two seconds. In a short time he regained his
former power, which he maintained for several minutes, and
then gradually lost it again. About five periods of alter-
nating loss and recovery took place in twelve minutes. By
variations in the methods of experimentation, the different
factors are eliminated, and he is able to conclude that the
centre of voluntary control is unaffected, but that this
periodicity is dependent upon " alterations which take place
in some of the mechanisms between the areas of the brain
originating the will impulses and the centrifugal nerves.' ? Dr.
Hodge 2 found that when he stimulated the spinal ganglia of a
cat continuously with an interrupted current, no change
of the cell took place. When he applied his interrupted
current for a quarter of a second and allowed the cell to rest
three-quarters, a change took place in the nucleus of the cell.
These experiments are inconclusive, as in the first case the
animal was given curari and in the second it was not. Dr.
Burgerstein 3 tested a number of school children by their
ability to multiply and add figures for four successive periods
of ten minutes, with five minutes' interval between the
periods of work. During the third period there was a
marked falling off in the amount of work accomplished and
an increase again during the fourth period. He argues that
the pupils became fatigued during the first two periods, and
that the third was a period of recovery, since the normal
amount of work was shown again in the fourth.
The secondary rhythm observed in the circulation occurs
also in respiration. Under ordinary circumstances respira-
tion follows a rhythm of about fifteen or twenty a minute.
During certain diseases and sleep a secondary rhythm —
Cheyne- Stokes 4 curves — appears in respiration. The respira-
tory movements decrease in depth until they disappear entirely.
After an interval of a few seconds a slight movement occurs.
This is followed by others, which increase in strength until
they become normal and sometimes abnormally strong. Two
explanations are offered : first, a waxing and waning in the
nutrition of the respiratory centres, and second, a rhythmic
1 Warren P. Lombard, M. D., "Effect of Fatigue upon Muscular
Contractions." Amer. Jour, of Psy. Vol. III.
2 " Microscopical Study of Changes due to Functional Activity in
Nerve Cells." Jour, of Morphology, Vol. VII.
3 Die Arbeitskurve einer Stunde. Zeitschr. f. Schulges. IV. 9, 10.
4 Foster's " Physiology," 6th Ed. p. 605.
152 BOLTON :
increase and decrease in the inhibitory impulses playing upon
the centres. The latter explanation is favored. This,
however, simply assigns the rhythmic action to some other
centre and does not explain the phenomenon. A certain
amount of secondary rhythm takes place in the breathing of
hibernating animals. Eespiration appears almost to cease
and then to start again, but it is generally slower during
hibernation.
Growth appears to take place rhythmically. Distinct
periods of activity and rest occur in the embryonic develop-
ment of some species that have been observed. This has
been seen in the segmentation of pulmonates' eggs. 1 It is no
less true of the amblystoma. In these the periods of activity
last from five to fifteen minutes, and are succeeded by inter-
vals of repose lasting about forty-five minutes. The activity
of the protoplasm offers a resistance which must be overcome
by the energy arising from the assimilation of the granular
food material, which disappears as development proceeds.
During the period of repose the energy is accumulating from
this assimilation, which, when it becomes sufficient, overcomes
the resistance, and activity sets in. This is taken to be a type
of physiological and nervous activity, which will serve to
explain certain phenomena of rhythm. This rhythm in
growth, which is observed in the embryonic development, is
characteristic of the physical and mental growth of children.
For several years previous to puberty, great increase in
stature is observed, puberty itself being a period of slow
growth. From fifteen to eighteen is another period of growth,
in which the full stature is generally reached. The mental
character of children shows also periods of activity and
repose. 2 The bright child becomes dull and the tidy slovenly.
The leader in the athletic sports is now lazy and moping.
Memory is now predominant, and now reason. The child
passes from one form of activity to another. The line of
development goes zigzag to its goal.
Other examples of involuntary action might be mentioned.
These are the peristaltic contractions of the intestines, labor
pains, the recurrence of heat and of the menstrual flow, and
the secretions of the digestive cell. In these cells the secre-
tions are kept up for about six hours, when a period of repose
of about twenty-four hours follows. 3
1 W. K. Brooks, "Fresh Water Pulmonates." Studies from
Biol. Lab. at Johns Hopkins Univ. Vol. II.
2 G. Siegert, "Die Periodicit&t in der Entwickelung des Kindes-
natur."
3 J. M. Langley and Sewell, "Histology and Physiology of Pepsin-
forming Glands.'' Phil. Trans. Vol. CLXXIL pp. 663-711. London,
1882.
RHYTHM. 153
From this review it may be safely said that nervous action
in general, and especially of the lower and vaso-motor centres,
is rhythmical. This form of activity results from the resist-
ance which the nervous substance offers to a stimulus. A
certain amount of energy is necessary to overcome this
resistance. This fact is brought out by the experiments of
Helmholtz 1 and Sterling upon the summation of stimuli.
Helmholtz found that when he stimulated a nerve going to a
muscle by a submaximal stimulus and then added another
stimulus at any time afterward within four seconds, he ob-
tained a contraction. If he used a maximal stimulus in the
first place and then added another stimulus during the latent
period, it produced no effect upon the contraction due to the
first stimulus. But if the second stimulus was added after
the latent period, the effect was a greater contraction than
that which followed the first stimulus alone. Submaximal
stimuli 2 following one another, even as slow as one per second,
will produce a contraction after a time. As the frequency of
the stimulus increases, the effect is much more marked. It
is much better to increase the frequency of the stimulus with-
out increasing the strength than to increase the strength
alone. Sterling adds further that all muscular and nervous
action is due to summated stimuli — a conclusion that denies
the possibility of contractions due to one instantaneous shock
or at least does not explain them. Dr. Ward 3 determined
that between the rates of .4 sec. and .03 sec. a contraction
always followed a given number of stimuli. Above and be-
low these limits the number might vary. In the same line is
the work of Drs. Kronacker and Hall.
It has been held by Sterling and others that when a stimu-
lus is applied directly to the cortex, no matter what the rate,
the brain sent out rhythmic impulses always at a constant
rate. Dr. Limbeck 4 conducted a series of experiments upon
the brain and spinal cord, in which he finds that the brain
and spinal cord send out just as many impulses as they
receive. Faster rates than thirteen shocks per second for
the cortex and thirty-four for the cord, gave smooth curves.
1 Helmholtz, "Berichte der Berliner Akad." 1854, p. 358.
2 W. Sterling, u Uber die Summation electrischer Hautreize."
Berlin. Berichte d. Sachsgelschaft d. Wissenschaft. December,
1874, p. 372.
3 Dr. Ward, " Uber die Auslosung von Reflexbewegungen durch
einer Summe schwacker Reize. " Archiv fur Anatomie und Physiol -
ogie. 1880, p. 72.
Hugo Kronacker und G. Stanley Hall. Die willkurliche Muske-
laction. Archiv fur Anatomie und Physiologie, 1879.
4 Dr. R. U. Limbeck, " Uber den Rhythmus centralie Reize. ,r
Archiv ftir experimentale Pathologic Bd. XXV. H. 2.
154 BOLTON :
The difference between the rates for cortex and cord is worthy
of note in consequence of the close correspondence of the
number of shocks for the cortex and the rate of the most
rapid voluntary control ; while involuntary and clonic con-
tractions which find their seats in the lower centres and in
the cord may be much faster. In this connection the atten-
tion of the reader is called to that portion of the experimental
study in which the rate of clicks at which rhythmical grouping
ceases is set forth. It is not far from ten a second. This is
also near the lowest rate at which air vibrations give the
impression of a musical tone.
The theory of summated stimuli which was advanced by
Wundt, and which is generally accepted, is based upon the
resistance which a central cell offers to a stimulus. The in-
coming stimulus is not communicated directly to the cell.
The afferent nerve does not terminate in the cell, but breaks
up into branches, which form a kind of envelope about the
cell. The efferent nerve takes its rise in the nucleus of the
cell and proceeds towards the periphery. If the stimulus is
weak, it does not penetrate through the surface of the cell to
the nucleus, but only part way. It sets up a kind of disturb-
ance around the surface of the cell and, should another stimulus
follow before the disturbance has subsided, it adds to the
effect already produced. Eepeated stimuli still further in-
crease the disturbance until it penetrate to the nucleus of
the cell, when it causes the cell to discharge into the efferent
nerve. This serves very well for summated stimuli, but
other phenomena of just the opposite nature require ex-
planation. There are the soothing effects of slow and gentle
stroking or patting, such as hypnotizers and nurses use upon
their subjects. The general fact seems to be that the stimu-
lus must not rise much above the threshold, and be sufficiently
slow, that there shall be no summation. As we shall see later,
any repeated stimulus tends to take the form of a muscular
movement accompanying it. If this stimulus becomes
gradually slower, it leads finally to the concept of rest, and
being accompanied by muscular movements, these movements
must finally cease. Increased quietude follows the slowly
decreasing movements, until before a great interval of time
has elapsed the body falls into a state of rest. The stimulus
must in any case be sufficient to command the attention of
the subject to the exclusion of the disturbing effects of other
stimuli coming from without and from the involuntary pro-
cesses of the body. Let us return now to the nerve cell, to
find if there are any processes going on which will throw
light upon the problem. A weak stimulus is continually
playing upon the cell from without, but never rises sufficiently
RHYTHM. 155
in strength to penetrate beyond the periphery of the cell or
in rapidity to bring abont a summation. The effect of each
stimulus subsides before the following one reaches the cell.
The peripheral area must soon become fatigued so that it is no
longer able to respond to the stimulus, and yet it is sufficiently
strong to command the attention in so far as to distract it
from other stimuli coming from within. This is the condition
of quietude in the cell which is manifest in the muscle.
Attention and Periodicity : — The most casual observer will
discover that his attention is discontinuous and intermittent.
It manifests itself in a wave-like form. It is a series of
pulses. The mind does not rest for any length of time upon
a single object. New phases and relations must continually
appear, or the object is dropped, that another may be taken
up. " No one can possibly attend continuously to an object
that does not change." 1 This process has been described as
a "fly and perch." Charles Pierce says in his "Philosophy
of Attention" that there is "no continuum." This periodicity
in attention has been observed by Helmholtz 2 with the
stereoscope and commented upon at considerable length.
The phenomenon is called retinal rivalry. Mr. T. Eeed 3 re-
cords some observations which he made in combining two
stereoscopic views, which were ruled, the one with vertical,
and the other with horizontal lines. He finds that the whole
field will be occupied for a time with one view, and then this
gives way for the other, which lasts an equal time. They seem
to change without voluntary effort and even in spite of one's
efforts to keep one view in the field. The full time for a
change from one to the other and back again is from twelve
to sixteen seconds for different subjects. The pulses of at-
tention, however, seem to succeed one another at much
shorter intervals. Two seconds seem a long time to hold
any object which has no relation before the attention. James
says : " There is no such thing as voluntary attention sus-
tained for more than a few seconds at a time." Does it not,
then, seem reasonable that during each wave or pulse of
attention only one undivided state of consciousness can arise?
The waxing and waning of attention seem to mark a change
from one object of consciousness to another. The object of
the state may be very complex, but it stands as a unit in
consciousness. The problem of the relation of the parts of
the object by which a great many may be allowed to stand as
1 James, " Psychology,' ' Vol. I. p. 420.
* " Physiologische Optik," Sec. 32.
»" Nature," August 1, 1887.
156 BOLTON :
a unit in consciousness and be grasped in a single state, is of
the most vital importance, but it must be deferred until later,
when the normal period of a wave of attention will also be
discussed.
Rhythmic Speech: — The most distinguishing, and in many
respects the most important, function of the human body is
vocal utterance and articulate speech. Being an involuntary
and habitual function in a large measure, it might be expected
upon a priori grounds to be rhythmical. Speech becomes
rhythmical not simply by sounds succeeded by pauses, but
also by the regular recurrence of strongly accented sounds in
$> series. Aside from the simplest shout or exclamation of
joy or pain, all vocal utterances are primarily rhythmical.
Every word that contains more than one syllable consists of
strong and weak syllables. These accents occur upon every
other syllable in varying intensity, or at most the accented
syllables are separated by two unaccented syllables. As
regards vocal utterances, they can be considered from four
different aspects — their regular succession, intensity, pitch
and quality. The problem in a philosophical treatment of
rhythmic speech is to determine the value of these properties
of sound as unifying elements in a rhythmical production.
It will be necessary first to inquire which is the most funda-
mental, and secondly, where each enters and the part it plays
in the development of literature. We must seek also other
unifying principles, if such there be. Of these, we might
now mention the logical meaning of words — the theme — and
aesthetic forms. As we are concerned in speech in so far only
as rhythmical effects are aimed at, we shall speak only of
poetry. By what coordinations and subordinations of sounds
with respect to their properties and meanings is the whole
structure of the poem held together ? It is the same problem
which Plato discussed as the one and the many. Kant put
the same question by asking how the mind made a unity out
of a manifold. We have to ask how the mental span becomes
so enormously increased as to grasp such a poem as Words-
worth's u Intimations of Immortality from the Eecollections
of Childhood," or Milton's " Paradise Lost." How is the
carrying power of the mind increased to such an extent?
The answer is to be found in the fact that unities are formed
out of the simplest elements of speech by coordinating some
with others in respect to their time relations; secondly,
unities are formed of unities by subordinating them with
respect to their intensities, and sometimes, their time values;
thirdly, by coordinations and subordinations with respect to
intensities and qualities, higher unities still are formed ; and
RHYTHM. 157
fourthly, by coordinations and subordinations with respect
to theme and aesthetic forms, the greatest unities are accom-
plished. In the first place vocal utterances are related as
regards time, that is, the same sound may recur at regular
intervals, in which case the series thus formed might be
termed a rhythmic series — a series which may become
rhythmical. In the next place this series might be made up of
louder and weaker sounds alternating with each other. The
series would then be composed of groups of sounds and might be
called a rhythmical series. This is a rhythm in speech. If
now the louder sounds in each group were given different
intensities, these smaller groups might be brought into larger
groups still. In this way the mental span may be made to
extend itself over a very large number of simple impressions.
The principle is very clear, and one will see at a glance that
if intelligible sounds were used and qualitative changes em-
ployed, the mental span might be almost indefinitely extended.
The carrying power of the mind, however, does not rest
wholly in any case upon a single fact, if we make the excep-
tion that vocal utterances must be carefully timed in a
rhythmic series. Quality and pitch changes accompany
changes in intensity, so that the subordination of one sound to
another and their consequent unification with respect to
intensity is always dependent upon pitch and quality changes
as well. For this reason it is impossible to treat each
properly by itself.
Time-relations : — In order for vocal utterances to form a
rhythmic series, they must occur at regular intervals of time
which cannot exceed or fall much below certain limits. We
may, however, upon the analogy of physiological rhythms, re-
gard a series of sounds recurring at stated intervals as a rhyth-
mical series, and also regard the recurrence of accented sounds
as forming a secondary rhythm out of the primary. This is
carrying the rhythmical idea farther than has been customary,
and while it is more nearly correct, it would not be generally
understood. The question of the time values of vocal utter-
ances for rhythmical purposes cannot be answered upon an
examination of poetry itself. Although the Greeks and
Eomans assigned exact values to all syllables in their lan-
guage, there is reason for believing that such values did not
arise naturally, but were assigned when they began to specu-
late upon poetry. No such relations exist among the sylla-
bles of modern languages, and in English they never did.
We must then dismiss the subject of time and its significance
and revert to it as the subject permits.
158 BOLTON :
Intensity of Sounds: — The mind accomplishes its first
real unification of sounds by subordinating them with respect
to their intensities. A rhythm in speech means a series of
groups of sounds. Each group may contain two or more
sounds, generally not more than four. Two sounds, one
strong and one weak, the one succeeding the other in time,
cannot give an idea of a rhythm, but two groups of two
such sounds certainly can. This being the simplest possible
rhythm, we should expect that it would be the earliest
form in which literature appeared. Since we have not
probably any extant specimens of the first literary productions,
for they were not committed to writing, we must judge from
those which have come down to us from later periods, and
from the literature of primitive peoples and of children,
what the earliest form was. In this way it has been proved
that our surmise, which was made upon a priori ground
simply, is correct. The oldest extant specimens of English
poetry are generally composed of verses of two sections,
which are separated by a pause in the middle. Each section
generally contains four, sometimes six, syllables, two of
which are unaccented and two accented. The first section
was emphatic and corresponded to the accented syllable in
the smaller division ; the second section received less stress
and was less important. The two formed a kind of balance
structure, in which the first section contained a rise and the
second a fall.
helle heafas: hearde nithas.
wer leas werod: waldend sende.
graes ungrande: gar secg theahte. 1
" Our Anglo-Saxon 2 poems consist of certain versicles, or,
as we have hitherto termed them, sections, bound together in
pairs by the laws of alliteration For the most part
these sections contain two or three accents, but some are
found containing four, or even five. The greater number of
these sections may be divided into two parts, which generally
fulfill all the conditions of an alliterative couplet "
These are the rules that Guest gives according to which the
elementary sections were constructed : 1. " Each couplet of
adjacent accents must be separated by one or two syllables
which are unaccented, but not by more than two." 2. "No
section can have more than three or less than two accents. ' 7 3.
"No section can begin or end with more than two unaccented
syllables." " When the accents of a section are separated
1 These lines are copied just as they appear in Guest's "His-
tory of English Rhythms," p. 189.
2 Guest's "History of English Rhythms," p. 158.
RHYTHM. 150
by two unaccented syllables, the rhythm has been called
triple measure ; and the common measure, when they are
only separated by a single syllable. JJ The greater propor-
tional number of accents makes the movement slower, and
adapts the measure for more solemn and graver subjects.
The triple measure is more suited to lighter themes. The
verse of the common measure is made more energetic by be-
ing begun and closed with accented syllables. They are
abrupt when too short, and become feeble when too long.
There was considerable variety of rhythm as early as the
fifth century, " as there certainly was in the seventh century,
when Csedmon wrote." 1 "It is, however, probable that the
rhythms were of a simpler and of a more uniform character. "
.... "Most of the alliterative couplets have only four
accents — very few, indeed, have so many as six."
The phenomenon of accompanying the changes of intensity
in a series of sounds with muscular contractions, led to the
early association of dancing with musical and poetical
recitation. Indeed, if we accept the current theory of the
origin of language as arising during the celebrations of
victory, dancing precedes even language. Just as an
animal jumps and frisks about as an expression of pleasure
at seeing his master, so our ancestors danced for joy over a
victory, or in the worship of their deity. They emitted cer-
tain vocal utterances in company with the tramping of the
feet, which in time came to have definite meanings and also
took on the rhythm of the dance. This rhythm was scarcely
more than the simple swaying of the body or the lifting of
one foot and now the other. Variations in the dance might
occur either in taking several steps forward and then several
backward, or to the right and to the left. These variations
would produce corresponding effects in the vocal accompani-
ment. The step of one foot would be stronger and a more
intense sound made to correspond to it. In the same way
either the forward or backward movement would become the
more important and give rise to the distinction of thesis and
arsis of the verse. Further groupings of the verses might
take place in the same way. The two-rhythm was appar-
ently the prevailing rhythm in the history of our language, if
not in some others. The most common foot in our literature
of all times, and a very common foot in the Greek literature,
consisted of two syllables ; two feet entered into the section,
and two sections formed an alliterated couplet or verse. It
is the simplest possible rhythm, and corresponds to the leg-
pendulum with which the language was so intimately asso-
ciated in its earlier history.
1 Guest's "History of English Bhythms," p. 169. *
160 BOLTON :
Noire 1 believes that language took its rise in the concerted
action of many persons. In this way the individual finds
that what belongs to him is the common character of others.
Such utterances as "hi-ho" are taken to be the first begin-
nings of language, and they originate during concerted action.
Any sound that is to become intelligible must first be ex-
perienced in company and then by the individual alone. But,
as the example shows, such utterances are rhythmical.
Here it is the rhythm of heaving sails or anchor, which is
seen among sailors.
Variations in the number of syllables to the accent would
be a necessity as a relief from the monotony of two syllables
to the accent, and so, too, the number of accents to the sec-
tion would be increased on account of the abruptness of the
doubly accented section. Taine 2 in speaking of early Saxon
poets says: "His chief care is to abridge, to imprison
thought in a kind of mutilated cry." "They (Saxons) do
not speak, they sing or rather shout. Each little verse is an
acclamation which breaks forth like a growl. Their strong
breasts heave with a groan of anger or of enthusiasm. A
vehement or indistinct phrase or expression rises suddenly,
almost in spite of them, to their lips." After the people be-
came settled down in their new homes, they lost the ruder and
rougher characteristics, and such wild outpourings would be
no longer suited to their milder spirits. The changes that
took place in the development of our literature are due in
some measure to the change in the life and habits of the
people.
There still remain in our poetical compositions certain
evidences of some, at least, of the stages through which our
poetry has passed. The choruses in many of our hymns are
still made up of non- sense syllables. Irish melodies and
popular songs retain this feature. Children's poetry — by
that I refer to such poetry as they enjoy and recite for their
own amusement — has a large element of purely unmeaning
sounds in it. Savage dances are often accompanied by
recitations in which no meaning has been discovered. Again,
savages and children are frequently found repeating for their
own amusement a series of non- sense syllables in rhythmical
form. The accents are very strongly marked, and frequently
enforced by alliteration. The incoherent chatter of a maniac,
or the sound of a foreigner speaking his language to one who
is unacquainted with the language, is distinctly rhythmical.
It is more like a chant, and children frequently remark upon
1 Ludwig Noire, " Max Miiller and the Philosophy of Language."
2 Taine, " Introduction to the History of English Language."
RHYTHM. 161
it. It appears, then, that vocal utterances which are kept
up for a considerable time fall into a rhythmical form. Such
being the natural tendency of speech, it would conform itself
to any rhythm with which it might be associated, and as vocal
utterances were always accompanied by the dance, it would
take on the rhythm of the dance, which in its earliest forms
we have seen reason to believe was the leg-pendulum.
The poetry of children shows a character very similar to
early English poetry. It consists often of a two-section verse
which is strongly alliterated, and in which the rhythm is
perfectly clear. The familiar incantation rhyme shows this
characteristic very well.
Sticks and stones
May break my bones,
But names will never hurt me.
Again,
Jack and Jill
Went up the hill
To fetch a pail of water ;
Jack fell down
And broke his crown
And Jill came tumbling after.
At the beginning of each couplet there is wanting one
syllable. Their poetry is usually accompanied by marching
or by clapping of the hands, so that they require an accented
syllable at the beginning. The verse is, then, an alternation
of accented and unaccented syllables ; occasionally only two
unaccented syllables occur between two accents. We have
in the first couplet what was found to be a prominent
characteristic of early English poetry.
E. B. Taylor in his " Anthropology" asserts that while meter,
and by that he means lines regularly measured in syllables, is
an evidence of civilization, one of its earliest developments is
matched and balanced sounds. The Australian savage sings
at the end of his verse, "A bang! A bang!" Certain of the
North American Indians sing in choruses, "Nyah eh wa!
Nyah eh wa!" The chorus of a New Zealand song is
u Ha-ah, ha-ah, ha-ah, ha!" A feature extremely common
in barbaric song is a refrain of generally meaningless sylla-
bles. Guest, 1 speaking of our early poetry, says, "I have
hazarded the opinion that these short, abrupt and forcible
rhythms were the earliest that were known to our language.
They are such as would naturally be prompted by excited
feeling, and well fitted for those lyrical outpourings which
form the earliest poetry of all languages." The abruptness
Quest's "History of English Bhythms," p. 365.
2
162 BOLTON :
is felt by children, so that not more than a single couplet
appears without the intervention of a different kind of verse.
Shakespeare 1 adopted this measure in his descriptions of
fairyland, and it is now become the fairy dialect of the
English language.
Qualities of Sounds : — Qualities of sounds are quite as
important as unifying elements as their time and intensity
relations, and were quite as early regarded. This is mani-
fest from the frequent recurrence of the same sound at the
beginning of Anglo-Saxon and Germanic verses. This is
alliteration. The two sections of the verse, while contrasted
in intensity, were coordinated by the recurrence of the same
sound. The origin of alliteration is involved in some
mystery, and yet the savage shouts just quoted point out a
possible origin. The emotional shout of an animal for a
given state is always the same ; but for the savage, who
possesses greater powers of utterance, emotions find various
expressions, or at least, if the expression begins with the
same sound, it ends differently. Although the New Zealand
savage shouts "Ha-ah" several times in succession, he
closes with " Ha!" When the child torments his companion
in the midst of misfortune, he says " Goody, goody gout."
Other expressions of a similar character, but used with a
different purpose, are "higelty, pigelty," "hee-ho," etc.
In modern poetry alliteration has given place in a very large
measure to final rhyme, which has become the unifying factor
for the verse generally in English poetry and always in
French. The qualities of sounds gave rise to melody in
speech, which is common to both poetry and music, and it is
as melody that the qualities of sounds play the most impor-
tant part.
Spencer holds, in his essay upon the origin of music, that
different emotional states produce different intonations and
changes in pitch, quality and loudness of vocal utterances.
In the savage dances of victory, worship, and love, emotional
speech grew up, and from this music arose. Originally
music was recitative — a mere chant. Chinese and Hindoo
music is still so. This recitative speaking grew "naturally
out of the modulations and cadences of strong feeling. " The
Quaker preacher who speaks only when moved by religious
emotion, speaks with a recitative intonation, and church
services of the present day are generally read so. This is
really melody. Eecitatiye speaking, or emotional speech,
constitutes the whole of savage poetry.
Poetry and music among primitive peoples were the same.
Poetry was either sung or chanted, and it was not until a
1 Guest's " History of English Rhythms," p. 179.
KHYTHM. 163
later period that they became separated. With the discovery
of the musical instrument, the people saw that a melody was
just as well expressed by simple tones as by intelligible
syllables, and music took up its own lines of development.
The Emotional Effects of Rhythm upon Savages and
Children : — There is no more striking fact in the whole field
of rhythm than the emotional effect which rhythms produce
upon certain classes of people, savages and children. Atten-
tion has already been called to the psychological phenomenon
of accompanying the changes of intensity in a series of
sounds by muscular movements. So strong is this impulse
in all classes of people that no one is able to listen to music
in which the rhythm is strong and clear without making
some kind of muscular movements. With some people
these movements tend to increase in force until the whole
body becomes involved and moves with the rhythm. The
accents in the rhythm have the effect of summated stimuli,
and the excitement may increase even to a state of ecstasy
and catalepsy. Although the regular recurrence of the ac-
cented syllable is the most important element, the qualita-
tive changes aid in bringing about the emotional states.
Soothing effects result from certain rhythms, as is shown in
the lulling and patting of the baby to sleep. The early
hypnotizers resorted to the gentle stroking of their subjects.
Savages are well aware of the exciting effects of certain
rhythms, and are accustomed to use them to bring about the
state of frenzy in which their priests give their prophecies
and in which religious dances are danced. Mr. Ellis, l who
has made a study of some tribes in Africa, says, " Music
amongst the Thsi-speaking tribes is limited to airs possess-
ing an obvious rhythm. Such airs seem to appeal to the
primitive sense common to all people, but upon savages, that
is, upon children with the possession and power of men, its
influence is immense, and the state of excitement into which
an assemblage of uncivilized people may be wrought by the
mere rhythm of drums and the repetition of a simple melody
would hardly be credited With some races this known
emotional influence of music has been utilized with three
objects, viz., to stimulate the religious sentiments, the mar-
tial spirit, and the sexual passions. "
In the Yatiati 2 dance among the Indians of British Colum-
bia, the tribe assembles outside of the chief's house in which
the dance is to be held, and with fists and sticks they beat
the time on the walls as they enter, singing the dancing
1 A. B. Ellis, "The Thsi-Speaking Peoples of the Gold Coast of
West Africa," p. 325.
3 Franz Boas, Jour, of Amer. Folk-lore, Vol. I. p. 49.
164 BOLTON :
song. The dancers who are on the inside are worked up
into a frenzy. The gentle striking at first, gradually increas-
ing in violence, and the slow approach and the assemblage of
the tribe, wrought in the dancers a pitch of excitement which
forced them to rush out after a time and begin the dance,
jumping about in the wildest fashion. Such dances cease
only with the complete exhaustion of the dancers.
The Patagonian wizard 1 begins his performance with drum-
ming and rattling, and keeps it up till the real or pretended
epileptic fit comes on by a demon entering him. Among the
wild Yeddas of Ceylon the devil dancers have to work them-
selves into paroxysms to gain the inspiration whereby they
profess to cure their patients. With the furious dancing to
music and the chanting of attendants, the Bodo priests bring
on a fit of maniacal possession. The excitement is allowed
to continue until the prophet falls to the ground in a
swoon. When the Alfurus of the Celebes invite their deity
to descend among them, the priests, standing about the chief
priest, upon whom the deity is to descend, chant some
legends. A slight twitching of the limbs marks the beginning
of the possession. The priest turns his face towards heaven,
the spirit descends upon him, and with terrible gestures he
springs upon a board and beats about with a bundle of leaves,
and leaps and dances, chanting some legends. He falls in a
swoon, and the sounds he emits are interpreted as the will of
the spirit.
George Catlin 2 says dancing is always accompanied by the
singing of mysterious songs and chants, which are perfectly
measured and sung in exact time to the beat of the drum,
always with an invariable set of sounds and expressions.
The religious services and singing among the Shakers are
often accompanied by dancing, and more frequently by beat-
ing of the time by all the members of the congregation. The
excitement among them never rises to an extreme degree. A
highly civilized people is not easily affected by mere rhythms.
A simple tone is not so expressive as it is to the lower
classes of people. The negro preacher often resorts to reci-
tative speaking to produce the desired emotional state in his
hearers, which is generally known as the "power." He selects
some short sentence, often unimportant, such as "Moses
went up into the mountain," and repeating this, at first
softly, he gradually raises his voice to the highest pitch, at
the same time increasing his gesticulations. The more ex-
citable of his audience are thrown into a paroxysm ; the con-
1 E. B. Taylor 2 " Primitive Culture."
2 George Catlm, " Letters and Notes upon the Manners and Cus-
toms of North American Indians.' *
RHYTHM. 165
tagion spreads so that sometimes the whole audience is in-
volved. Evangelists among all classes of people rely more or
less upon the emotional effect of rhythmical speaking.
Street hawkers and fakirs generally speak with a recitative
intonation. Their success depends very largely upon their
success in alluring and holding the attention of the crowd by
the manner and intonation with which they speak.
The effect of rhythm and clearly accented music is no
greater upon primitive peoples than upon children. Al-
though children are not allowed to go into ecstasies, the
clapping of the hands to the recitation of "Peas porridge
hot" is akin to the terrible leaping and gesticulations of the
savage to the accompanying tom-tom and the chanting of his
ancient legends. The child usually begins his recitation of
"Peas porridge hot" rather slowly, and as he continues he
grows in excitement and enthusiasm, his gestures become
more violent and rapid, until he breaks down in the excite-
ment. It is a well-known fact among school teachers that
young children become excited whenever they sing rhymes
with a strongly accented rhythm. Several have made this
observation during the singing of a certain line in Theodore
Tilton's "Baby Bye." The line in which the excitement
reaches its climax is,
There he goes
On his toes
Tickling baby's nose.
This is a type of the fairy measure. The accents are
strong, and every line is preceded by a pause, and at the
same time all the lines are rhymed. Both the rhyme and the
pause lend an intensification to the rhythm that is sufficient
to call out the greatest excitement in the fairy people. In
Eobert Browning's poem of the "Pied Piper of Hamlin,"
whose charm was rhythm, occurs this remarkably rhythmical
passage, and taken with the context might easily cause some
emotional excitement :
Into the streets the piper stept,
Smiling at first a little smile,
As if he knew what magic slept
In his quiet pipe the while.
I have the testimony of an eminent educator that, when he
read these lines, and he is an effective reader, his boy, a
youngster of five or six years, would run away and hide
where he could not hear the reading. He was apparently
unable to bear the strain of the excitement. In later years
the boy could not tell why he did so, except that it disturbed
him.
166 BOLTON :
The use by children of incantation rhymes for purposes of
injury and torment to their companions is interesting in this
connection. The habit of rhyming is almost instinctive with
them. Imagine the effect of such a couplet as this upon the
child to whom it is addressed :
Good night,
Rosie Wright.
Again, any name may be put in certain adaptive rhymes
which are current among children. These, however, are not
so effective as the instance cited above. They admit of
retort. The drawling out of a name in a sing-song measured
tone is very effective, and the easy adaptation of some names
makes the child who is unfortunate in having such a name an
object of torment.
The Place of Rhythm in Music and Poetry. Music : — We
have seen how music and poetry took their rise together
from the emotional utterances of savages during the dance,
and how these emotional utterances gradually took the form
of recitative speaking. This gave rise to the melody, though
it was not disassociated from the meaning of the words.
With the discovery of the musical instrument came the
discovery that a melody might be sustained by simple tone
intensities. Although music finds its essential basis in
rhythm, its distinctive feature is the melody combined with
harmony. The melody is constituted of a succession of tones
which are significant of an emotional state, and when several
melodies are combined and sung together, they give rise to
harmony. This combination of melodies depends upon the
pitch of the sounds. The melodies in harmony are all
subordinated in different degrees to one dominant melody
which is higher in pitch than the others. The unifying ele-
ment here is pitch. This is the only distinctive use that is
made of it in either music or poetry. The most important
and fundamental unifying principles underlying music is the
time, without which there can be no music. Musical tones
must be exactly timed, if one is to get the conception of a
melody from a series of tones. When they are exactly timed
they may be farther unified by regular changes of intensity
which group the sounds into measures. The most common
measures that occur in music are 2-4, 3-4, 4-4, and 6-8 time.
In what might be termed the natural system of accents, the
first note in each measure receives a strong accent. This is
really the only accent in 2-4 time. In 3-4 time the second
note also receives an accent, but it is weaker than the first.
In 4-4 time there are four grades of intensity. The first note
is the strongest, the third next, the second is weaker still,
RHYTHM. 167
and the fourth is the weakest of all. In 6-8 time the third,
fifth and sixth are of about equal intensity, and weak. The
first is strongest, the fourth is next, and the second weaker
though stronger than the third. An equal amount of time is
given to each measure — that is, the strong accent occurs at
regular intervals — but the distribution of this time among
the notes in a measure may be greatly varied ; the separate
notes, however, always bearing constant and simple relations
to one another. The smallest fraction that may express the
relations of these notes is 1-64, and this appears only in
instrumental music. In poetry, as we shall see, there is not
so much freedom ; it has deviated less from the primal
rhythmic stock from* which both spring. For many centuries
music consisted wholly of melodies, or of a single melody.
The idea of combining or singing several melodies at the same
time came very much later. This is harmony. It reached its
highest development about Elizabeth's time, when the at-
tempt was made to combine as many as forty melodies. A
much smaller number was found to give better effect, and the
number now used is generally only four. Symphony was a
still later development, but the general feeling among mu-
sicians now is that it culminated in Beethoven, and its
further development in music is impossible. Although the
term has had several significations in the history of music,
in Beethoven it was the combination of several themes in
such a way as to bring about a succession and combination of
strong emotional states. The musician who desires now to
produce new effects, turns to the Yolks-Lieder for a theme.
He aims at variations of the rhythmical effects and introduces
new harmonies. Mendelssohn is said to have remarked,
when he heard some of the negro melodies of our slaves, that
here was a field for a great musical talent. Wagner, taking
the suggestion, has made such an adaptation of the Hunga-
rian melodies, and with what success the musical world is well
aware. Wagner has made a real advance, and for some time
musical composition will follow his lead. Although
there is a feeling among musicians that rhythm is distasteful,
it is more apparent than real. It is the regular monotonous
recurrence of the same rhythm without sufficient variations
that is displeasing and not the rhythmic flow itself.
Rhythms in Poetry: — We have already seen that when
language appeared as literature, it took the form of the
simplest possible rhythm. Even then it was the vocal
accompaniment of a dance, and there are many analogies to the
simple swaying of the body or the tramping of the feet in the
march. There were no fixed rules in regard to the number
of syllables to the measure. The verse, so far as we can
168 BOLTON :
speak of a verse, consisted of an alternation of accented and
unaccented syllables. Very generally it began and ended
with an accented syllable, so that a pause occurred between
each verse. The line of development along which poetry
followed was an increase in the number of unaccented sylla-
bles as compared with the accented, and also an increase in
the number of accents to the verse ; the verse preserving for
some time the same balance of structure that it had in the
beginning. The number of accents then might be four, six
or eight; the latter number never became popular, for the
reason, it would seem, that it exceeded the normal mental span.
This even and balanced structure could not hold out forever ;
a demand for variety and the influence of foreign rhythms con-
tributed to overthrow it, so that Chaucer wrote altogether in a
verse of five accents, but he still retained the middle pause.
This came after the second accent or just before the third,
though sometimes after the third also. There were many
verses in which the first section more generally contained
three accents.
Guest takes no account of the measures or feet in English
verse. He divides lines into three general classes : Those
that begin with an accented syllable, those that begin with
one unaccented syllable, and those that begin with two unac-
cented syllables. The varieties in each of these classes
depend upon the position where the variation occurs from the
form in which the verse sets out. Should the verse begin
with an accented syllable and continue with an alternation of
accented and unaccented syllables, it would constitute one
variety. If, however, two unaccented syllables occur between
any two accents, it would constitute a different variety accord-
ing as t the two unaccented syllables occur between the first
and second accents, the second and third, and so on through
the verse. Early poetry was sung to the accompaniment of
the harp and hence was sung in exact time. On this account
Guest says that up to the fourth century, English rhythms
were temporal and then became accentual. Previous to that
time the syllable had a time value. This, however, is not to
be taken in any absolute sense. Poetry was chanted in a
kind of trance state, and the reciter aimed to produce such a
state in his audience. For this purpose the thought was of
minor importance. Great dependence was placed upon the
rhythmical flow, and doubtless a very exact time was given to
the syllables that the movement might be clearer. A rhythm
which depends wholly upon either the time element or the
accent, is certainly less forcible than one which combines both
factors. It must be conceded that though some regard was
paid to the time of syllables, no such exact time was main-
RHYTHM. 169
tained as modern musicians keep in their music. Perfect
time is the result of the application of scientific methods to
music. Poetry has never lost the time element entirely, for
accents that occur at irregular intervals could not have been but
very displeasing, and they are now. It is reported of some of
our modern poets, and especially of Tennyson, that they read
their poems with the strictest observance, not only of the ac-
cents, but of the time, showing that they regarded the time ele-
ment of great importance. Many readers and teachers of
English poetry pay little heed to the regular recurrence of the
accent. For them the thought is the chief element in poetry,
and in attempting to bring that out, they disregard the rhythmi-
cal flow. But when the proper observance of the thought does
violence to the rhythm, the poet must be adjudged lacking
poetic inspiration, and to that extent his poetry is not true
poetry. It is to the great renown of Chaucer, Milton and
Shakespeare that there is such a perfect adaptation of the
rhythm to the theme in hand, and any lack of observance of
the accents by the reader betrays his want of understanding
of that which he reads. The strict observance of time in
music and the unity of origin of poetry and music, which
argues that time was once an essential element of poetry, show
that the time element is still there, unless it can be shown
when and why it has dropped out. Poetry has admitted
fewer variations and allows a greater prominence to the
rhythmical flow than music. It must be admitted, however,
that the thought has taken the place of the melody to a
great extent as the unifying element, but it cannot be allowed
to take the place of other factors. Whenever it does, just so
soon the composition fails of being in any sense poetry.
Alliteration, which was very prominent in Anglo-Saxon,
was gradually lost. The influence of the church and of Latin
scholarship aided somewhat in this movement, but as the
Anglo-Saxon element prevailed against all foreign influences
in the political and social affairs, it won the day in the
struggle against the Norman and Latin languages. Our lan-
guage remains essentially Anglo-Saxon, and alliteration,
though less common, is still a prominent feature of our*
poetry. Originally, alliterated syllables marked the beginning
of the section and constituted the unifying factor of it, but
there was no strict observance of such a principle, except that
the alliterated syllables were accented. They might come
anywhere within the section. The use of alliteration by later
English poets was to place the alliterated syllables away from
the beginning of the section and to put them in the same verse.
The purpose of alliteration is not to coordinate two sections
or two lines, but, by intensifying certain accents in the verse,
170 BOLTON :
to make a more perfect subordination of them, or to make a
more perfect unity of the line. Final rhyme succeeded allit-
eration. The chief reason seems to have been for a more
emphatic or distinguishing mark of the rhythm than could be
obtained through accents alone; especially when run-on
lines came to be used and the thought was about to usurp
everything. When two successive sentences or words begin
with the same sound, it interferes with the understanding of
them. Both the reader and hearer are more likely to con-
found them. For this reason alliteration must give way,
except for purposes of emphasis, when the thought becomes
of the first importance. Simple intensities are not sufficient
as unifying factors ; they cannot be properly subordinated to
give unity to the line. It is interesting to note how the change
from alliteration to rhythm has come about. In the early poet-
ry, the alliterated syllables came at the beginning of the verse,
but in modern poetry the rhymed syllables, which are their
successors, come at the end. We shall see later how the begin-
ning and end of rhythmical groups run into one another
and become indistinguishable. The same is to be observed
with reference to the feet. The accents in the feet become
transposed. Although it seems probable that the foot in
early poetry and the measure in all music began with the
accented sound, the accented syllable in English poetry is
more generally the last, and in Latin and Greek poetry it was
quite as frequently the last as the first. The series of accented
syllables in the verse and of articulate sounds in the foot seem
to appear as a series of stimuli which are to be summated.
The two sections of the verse in old English were made to
rhyme with their last syllables, and were then written as two
verses. Two such couplets together form the most common
stanza in English poetry. Instead of writing the members of
each couplet next to each other, they are made more fre-
quently and quite generally to alternate.
JEsthetic Forms : — That which binds the four verses into a
stanza is not wholly the interelation and balance of the two
rhyming couplets. The members of the two couplets are fre-
quently made to begin, the one with an accented syllable and
the other with an unaccented syllable. Sometimes this, and
sometimes a less number of accented syllables, make the
lengths of the alternate lines less — a fact that gives artistic
form to the verse when it is properly printed. It becomes
then an appeal to the eye as an aesthetically beautiful form.
This principle was seized upon by our poets during the six-
teenth century, and carried to an extreme as regards form
alone, which could not be sustained by the thought. The
BHYTHM. 171
poem had nothing but form. The principle of form becoming
a unifying factor for a poem is perfectly true, and effective
use is made of it in modern poetry. Among the older poets,
George Herbert introduced many novelties into the forms of
stanza. He relied upon both rhymes and artistic forms.
Some of his stanzas take the form of a vase, an hour-glass,
a pyramid and an inverted cone. Although they read
smoothly, one cannot help but feel that his attempt at aesthetic
forms has destroyed the beauty of the poems.
The sonnet 1 is probably the most organic of all poems.
While the theme is very essential in binding the whole
together, the lines are coordinated in the most intricate way
by rhymes. A rhyme-scheme runs through the whole,
which, when represented by letters, or dots of different
sizes, or lines of different lengths, forms an artistic group,
obeying the laws of principality, subordination, etc. The
number of accents to the line is varied in some cases in such
a way that it lends a kind of subordination of some lines to
others, or of all to one or two.
Theme : — Little or no regard is paid to the thought in a
poetical recitation by children or by primitive peoples. They
delight in the emotional effect of sounds properly measured
and balanced. With the growth of literature the thought has
gradually become more important until it is about to usurp
everything. The unity of the stanza and of the verse very
generally depends upon it. Higher unities of the stanza —
poems— depend entirely upon the theme. The attempts to
coordinate stanzas by rhyming their last lines have not
proved a great success. The strength of the connection is
often lost. If the thought in the verse or stanza is allowed
to become the prevailing element, the poetry becomes
measured prose. Poetry arose in a kind of trance or highly
emotional state, and for centuries it was used to produce
such states in others. The whole structure is calculated to
produce emotion, and for that reason it cannot easily become
the medium of expression for the intellect. There must be a
mutual dependence between the thought and the form, or
they result in mutual destruction.
Under the influence of the church and Latin scholarship,
English poetry became, or at least the attempt was made to
conform it to certain rules of Latin prosody. English
critics, misunderstanding probably both English and Latin
poetry, tried to make the former conform to the rules of the
latter. And there are many persons now who cannot see
1 The reader is referred to Prof. Corson's " Primer of English
Verse " for a treatment of the stanza and sonnet.
172 BOLTON :
why the rules of Latin prosody are not universal. A verse
beginning with an accented syllable and consisting of an
alternation of accented and unaccented syllables, was trochaic
measure, and the accented syllable was double the length of
the unaccented. If two unaccented syllables were used be-
tween two accents, it was the substitution of a dactyl for a
trochee. In this case, if the syllables preserved their proper
time values according to the Latin prosody, four time-units —
the dactyl — would appear in the place of three — the trochee.
This, however, did not strike the critics as forming a defect
in the rhythm, and the error has gone on. It is the current
view among respectable English authorities to-day. In order
not to keep the reader in suspense about so disputed and
important a point, let me say what seems to be the true view.
As the simplest time-unit of Greek poetry was a short
syllable, and whatever value in time was given to it in a
verse, that value must be maintained throughout, so the
simplest unit of English poetry is the time between two
accents — the foot is the simplest unit in the verse — and this
must be constant. The time is apportioned among the
syllables that are present between the accents, whatever the
number. From the very nature of the accent the syllable
receiving it will be longer relatively, though it does not bear
a constant and simple relation to the length of the unaccented
syllables. " Besides the increase 1 of loudness and the
sharper tone which distinguishes the accented syllable, there
is also a tendency to dwell upon it, or, in other words, to
lengthen the quantity. We cannot increase the loudness or
the sharpness of the tone without a certain degree of muscu-
lar action ; and to put muscular action into motion requires
time."
Another fact which has been greatly overlooked in the
study of English rhythms, and which has led to much con-
fusion and erroneous speculation and criticism of some poets,
is the sectional pause, which allows two accented syllables to
stand together in the verse. It was very common in Anglo-
Saxon poetry, and disappeared almost entirely under the
influences spoken of above. Shakespeare made free use of it,
and for a lack of this knowledge, critics assert that he made
use of false accents. Caedmon placed it before words upon
which he desired to have a strong emphasis. It occurs be-
fore names of the deity. Guest says it owes its existence to
the " emphatic stop," and is really the greatest departure
from the rules of accent, which were observed with much
care by the Saxon poets. It has been revived by more
1 Guest's " History of English Rhythm," p. 75.
RHYTHM. 173
recent poets, and effective use is made of it. The value of
this pause is the great emphasis it lends to the word follow-
ing, and my purpose in dwelling upon it now is that it will
come up again prominently in the experimental study.
Another question which connects itself very closely with
this point of accent and pauses, is the foot or measure
division of the line. Guest does not recognize such a
division as the foot. The line is an alternation of accented
and unaccented syllables, and he does not mention the fact of
these forming groups which in Greek prosody were called
feet and in music are termed measures. There seems to be
no question that readers do make such groups by placing a
slight pause either after or before the accented syllables.
The Greeks associated these groups with a complete step in
the march, and since in matters of aesthetics it is a rash thing
to dispute or deny the accuracy of their judgments, we must
regard the foot as a real division of their verse and inquire
whether the lack of quantity in English syllables has any-
thing to do with the absence of the foot division. The
English verse is made up of a series of syllables in which
every other one is uttered with greater intensity than the
rest. The accented syllable requires more time, and the
unaccented syllable unites or fuses with it into an organic
group. These groups are then apparently separated by
pauses. In French poetry there are no accented syllables,
and the foot division is not recognized at all. This gives
English poetry a kind of intermediate position between
Greek and French poetry. The question of a foot division
cannot be finally answered from an examination of our
poetry, except as has already been said, such divisions are
invariably made. The question will find its final answer in
the experimental investigation.
Another problem which follows closely upon this is, what
is the inherent nature of a group in a rhythmical series, or,
what is the relation of the different syllables to one another
in the poetical foot, and what determines the length of it?
The length of sentence 1 in prose is found not to deviate
long from an average. Long sentences may prevail in an
author for a few pages, but they are sure to be followed by
short ones in sufficient number to balance the long ones.
There appears a kind of rhythm in which long and short
sentences succeed one another. This rhythm is constant for
the same author ; his earlier and later writings show no
difference in the length of sentences. The writers of the
more ancient prose show a greater average length of sentence
than our more recent writers.
X L. A. Sehurmann. University Studies, Nebraska University,
Vol.1.
174 BOLTON :
There have been several attempts in late years to con-
struct philosophies of English verse. Several of these will
be taken up and their more salient features presented. The
purpose is not to give a complete review of the books, but to
<?all attention to a few facts which will supplement the work
that has gone before.
The Science of English Verse by Sidney Lanier: — A simple
auditory impression recurring at regular intervals of time
furnishes the essential conditions of a rhythm. Of the four
properties of sounds — duration, intensity, pitch, and tone-
color — the mind can and does form exact coordinations of
duration, pitch, and tone-color ; intensities cannot be com-
pared with exactness. The regular recurrence of sounds and
silences constitutes primary rhythm, and a grouping of these
sounds by means of intensity, pitch, or tone-color, consti-
tutes secondary rhythm — the bar in music and the foot in
poetry. For purposes of verse, syllables correspond to
sounds and bear relations to one another in point of time,
which are expressed by the simple numbers 1, 2, 3, 4, etc.
The regularly recurring syllables of a sentence, whether prose
or poetry, constitute a primary rhythm, " which the
rhythmic sense of man tends to mould into a more definite,
more strongly marked and more complex form, that may be
called secondary rhythm. " " The tendency to arrange any
primary units of rhythm into groups, or secondary units of
rhythm, is so strong in ordinary persons that the imagination
will even affect such a grouping when the sounds themselves
do not present means for doing it." Accent simply
arranges the " materials already rhythmical through some
temporal recurrence. " As the comprehension of a series of
sounds is rendered more easy by grouping, so the compre-
hension of a series of these groups is rendered more easy by
again grouping these groups into tertiary rhythms. Allitera-
tion, the recurrence of emphatic words and punctuation
marks signify the tertiary group. The fourth order of rhyth-
mical grouping is the line which, except in the case of run-
on lines, completes a logical division of the sentence. Lines
are again grouped into couplets by tone- color coordinations.
The fifth order of rhythmical grouping is the stanza, and a
complete poem is spoken of as the sixth order.
The effort of the author, in his treatment of the foot, is to
make the rhythmical accent and grouping correspond to the
logical accent and meaning. For this purpose he treats at
length the iambic foot, it being the most common in English
poetry. Making use of musical terms, this foot is equivalent
to three eighth notes, and its typic form is one eighth note
RHYTHM. 175
followed by a quarter note. Instead of the eighth note, the
foot may contain two sixteenth notes, and instead of a quarter
note, there may be two eighth notes, or a dotted eighth and a
sixteenth. The foot may also contain three syllables, each
being equivalent to one eighth note, or four syllables, but
the four must be read in the time of three eighth notes.
In the place of any note, may be substituted a rest of equal
length. An anapaest or dactyl cannot take the place of ar
iambic or trochaic foot, since the former are equivalent to foui
time-units and the latter to three. He says there are two kinds
of rhythm only — 3-rhythm and 4-rhythm. All other kinds
resolve themselves into these two ; 2-rhythm is really
4-rhythm, and 5-rhythm is equivalent to a 3-rhythm and a
2-rhythm combined.
A Primer of English Verse by Hiram Corson ; — The
object of verse to him is "the expression of impassioned
and spiritualized thought. " It originates in "the unify-
ing activity of feeling and emotion. 77 Upon whatever
objects " feeling 1 or emotion is projected, or with what it is
incorporated — it is unifying." "The insulated intellect, in
its action, tends in an opposite direction — that is, in an analytic
direction. When feeling is embodied in speech, that speech
is worked up ... . into unities of various kinds." The primal
unity is the foot, which is combined " in a still higher unity
which is called the verse, and this in turn is combined into a
still higher unity, which is called the stanza." " Rhythm is
a succession and involution of unities, that is, unities within
unities." It applies to a succession of either feet, verses or
stanzas. Each class of unities has its combining principles ;
that of the foot is accent. Melody is the combining principle
for the syllables. Alliteration is a common and effective
form of consonantal melody. The combining agencies of the
stanza are harmony and rhyme. Individual verses may
be melodious, but when several are taken together they lack
harmony. Rhyme is also an enforcing agency of the individ-
ual verse, and the emphasis resulting is neutralized in pro-
portion as the verses are separated. Blank verse depends
"upon the melodious movement of the individual verses,
pause melody, and the general harmony or toning." Varia-
tions of the theme-meter produce important effects. "The
feelings of the reader of English poetry get to be set, so to
speak, to the pentameter measure, as in that measure the
1 This is the author's great mistake. No such distinction can be
drawn between feeling as unifying and intellect as analytic. Both
analysis and synthesis are equally properties of the intellect, and
it is difficult to conceive how the feelings can accomplish a synthe-
sis or unify anything.
176 BOLTON :
largest portion of English poetry is written." The intro-
duction of any other than the theme-meter gives an emphasis*
to the thought. The substitution of a different foot gives a
variety " which is essential to harmony." The shifting of
the regular accent gives a special enforcement, either logical
or aesthetic. " There should never be a non-significant
departure from a pure monotony."
Rationale of English Verse byE. A. Poe: — Verse originates
with the human enjoyment of equality. Unpracticed ears ap-
preciate simple equalities. Practiced ears appreciate equalities
among equalities ; they are able to compare two sets of equals.
The rudiments of all verse may possibly be found in the spon-
dee. In this, the mind finds its first pleasure in the equality
of two accented syllables. A collection of two spondees — two
words of two equal syllables — forms the second step in the
development of the verse. A third step would be found in
the juxtaposition of three words. This, however, gives the
idea of a monotone, a relief from which is found in words
of different accents — iambics. A dactyl might be employed
as a further relief from the monotone. A sequence of words
of any sort would form a monotone, if they were not curtailed
or defined within certain limits. This gave rise to the lines,
the terminations of which are again determined by equalities
in length, and marked by equalities — likenesses in sound.
Every foot in the same verse requires equal time. A three
syllable word may appear as iambic or trochee, providing
that two syllables can be read in the time of one. Blending
is an unwarranted liberty. He states this general principle :
The substitution of a foot, the sum of whose syllabic times is
equal to the sum of the syllabic times of the foot substituted,
is allowed with this restriction only, that the regular foot
shall continue long enough or be sufficiently prominent to
leave no doubt of the kind of verse. He says " that rhythm
is erroneous, which any ordinary reader can, without design,
read improperly." The real test of the perfection of a verse
is the pleasurable feeling it yields.
Classical Poetry: — Classical Greek poetry was either
chanted or sung, and for that reason was exactly timed.
There was really no difference between a poetical recitation
and a song. The simplest elements in the measure, accord-
ing to which poetry was sung, was a time-unit equivalent to
one eighth note. By combining these time-units into groups,
they formed the measure or foot. A group of several feet
constituted the section, and two sections entered into the line,
a certain number of which were united into strophes or
stanzas. A time value was given to all syllables and words
RHYTHM. 177
in the language ; they were either long or short. A short
syllable was equivalent to one time-unit, and a long to two.
Various measures were employed. They might be equal to
two, three, four, five or six time-units. The most common
measures contain three or four time-units. The three time-
unit foot most generally contains two syllables, one long and
one short, or one short and one long. The four time-unit
foot contains two or three syllables, generally two long, or
one long and two short, or two short and one long.
When only one long syllable occurred in the foot, it received
an accent ; when there were two in the foot, the first received
the accent. The accented portion stood as the thesis, and
the unaccented as the arsis. In the same way the two sec-
tions of the verse stood as thesis and arsis. The thesis came
first. The middle pause did not usually divide the verse
into two equal divisions. The first was the shorter, the
pause coming within the third foot. Except as showing a
perfect subordination to a chief accent, and a slight anacrusis
at the close, the verses had no distinguishing marks ; they
were not rhymed, and very rarely alliterated.
The number of feet in a verse varied with different kinds
of poetry, two being the smallest and six the greatest.
The kind of foot with which the measure set out was
not always maintained. Any other foot agreeing with the
theme-foot in position of accent, and in the number of time-
units, might be substituted. As such agreements in the
kind of feet were few, there could be very little variety in the
verse.
Greek poetry was not allowed to develop long untram-
meled by rules. A rigid philosophical system was imposed
upon it, and all future poetry was made to conform to this
system. But it would be difficult to say that Greek poetry
suffered from the restriction. It prevented novelty for
novelty's sake, but allowed great freedom where freedom was
most needed.
There are several facts in the history of rhythm that are
interesting, both for the subject in hand and for psychology
in general. Soon after the idea of varying the number of
syllables in a foot had become known, and its effects appre-
ciated, there arose a kind of mania for verses which contained
a variety of feet. They were characterized as " tumbling
verses " from the peculiar effect they gave rise to. This was
a discordant and unpleasurable feeling. There was really no
rhythm to them, and they never became popular. The same
took place in regard to the length of line. Various novelties
were introduced, when a longer line than that of the earliest
poetry was found more pleasing and less abrupt. Verses of
178 BOLTON :
six and seven accents were tried, and verses containing two
sections, each of which was an alliterated couplet, having
four or six accents, appear in some authors. No new com-
bining agency was employed, and probably for that reason
the verses exceeded the mental span. Had the older poets
grasped the principles of unifying their lines by rhyme, or by
proper subordination of the sections, they might have made
such long verses a success. In the same line were the
attempts at aesthetic forms, which have already been spoken
of.
Experimental Investigation.
This work was undertaken with several objects in view.
The first and most important object was to determine what
the mind did with a series of simple auditory impressions in
which there was absolutely no change of intensity, pitch,
quality or time- interval. Each separate impression was to
be indistinguishable from any or all the others. Eegular
variations with respect to the intensity or time-interval of
the sounds in this series, which will be called a rhythmic
series, were then to be tried separately and together, with
the purpose of determining what values these properties of
sound have in forming a rhythmical series — that is, a series
of groups of impressions — out of a rhythmic series. It was
seen at the outset that it would be practically impossible with
the apparatus at our disposal to employ pitch variations, and
for that reason no attempts were made with variations in
pitch. Yariations in quality or tone-color were contemplated,
but the experiment was not carried out, first on account of a
lack of time, and secondly of proper apparatus. The results
of the first experiment anticipated much that was to be tried
in the later experiments. As the work progressed, new
problems were suggested for investigation until the narrow
limits within which the work was begun were greatly over-
stepped. These problems will be taken up in what seems to
be their proper order, and the results presented.
Apparatus : — The click of an electric telephone when con-
nected in an induction circuit is constant in intensity, pitch
and quality, when breaks occur in the primary circuit, pro-
viding the primary circuit is constant. The click is not the
same in intensity when the primary circuit is made as it is
when the primary circuit is broken. For this reason, the
sound at the break only could be utilized. It is perfectly
constant and stronger in intensity than the click at the make.
It varies directly in intensity with variations in the strength
RHYTHM.
179
of the current and changes slightly in pitch and quality with
variations of intensity, but the pitch and quality are always
the same with the same intensity of current. A break at
regular intervals in the primary circuit, when the secondary
circuit is closed, the secondary circuit being open when the
primary was closed, was all that was necessary to furnish
the required series of auditory impressions with which the
investigation might begin.
A chronograph after the pattern devised by Wundt 1 and
Figure I.
built by C. Krille, furnished a constant power. Figure L
gives a general view of the whole apparatus as it was used in
this experimental investigation.
The drum-shaft was slipped off the drum and five arms,
two and one-half inches long were put upon it by passing the
shaft through a hole near one end. Each arm was provided
with a set screw, that the arm might be held in position and
ir This apparatus will be found fully described in the second volume
of Wundt's Physiologische Psychologie, p. 279. 3d ed.
180
BOLTON
its position changed at will. They were set at equal distances
apart along the shaft, and their points separated by 72
degrees, so that the space about the shaft was divided equally
into five divisions. (See Figure II.) Corresponding to
Figure II.
H'
Figure II. shows the operations of the keys.
A. Drum -shaft.
B. The wooden arm on the drum- shaft.
C. The dotted line represents the following point of the arm.
D. The continuous line represents the leading point.
E. The dotted line indicates the position of the arm when the
key is pressed down.
F. The key bearing the platinum points, which project below
and are connected by the wire indicated by K. The dotted line
below shows the position of the key when the platinum points dip
in the mercury.
H. The rubber elastic which caused the key to react.
I. The rod upon which the key turned.
L. The mercury cup.
M. The wire connections.
each arm were two keys placed in such a position on the top
of the chronograph that as the shaft revolved the ends of the
arms came in contact with the ends of the keys and pressed
them "downward about half an inch to allow the arm to pass
by in its revolution. The keys, which were ten in number,
two to each arm, were made of strips of wood, six inches
long and a half inch wide, and hinged horizontally upon a
steel rod two inches from one end in such a way that the ends
might move up and down. To the short ends were attached
elastics, which caused the long ends with which the arms
came in contact to rise up after they had been released by
the arms on the drum- shaft. They were prevented from rising
up too far by a piece of wood placed above them. Each arm
BHYTHM.
181
bore two points, the one about an inch to one side and ten
degrees in advance of the other. The leading point came in
contact with one key and pressed it down in advance of the
other. As each point was broad, covering about twenty
degrees of the circle described by the end of the arm, the first
key would remain down until after the other had been pressed
down. As both points upon each arm were of the same width,
the key first pressed down would be released before the other.
Eear the long end, each key carried two platinum points
which projected downward below the key, and which were
connected at the upper ends by a wire. When the keys came
down, the platinum points dipped into cups of mercury, which
rested upon the top of the chronograph. ( See Figure II.)
These mercury cups were made by boring holes into the side
Figure III.
6 6.
Q Q Q-
Figure III. shows the electrical connections.
A. The strip of hard rubber.
B. The battery.
C. The mercury cups.
E. 1, 2, 3, 4, 5. The primary coils. The double rings about 1
represent the induction coil.
K. The key-board.
The primary circuit is represented by light lines indicated by Y,
and the secondary circuit by a heavy line, X.
T. The telephone.
182 BOLTON :
of a strip of hard rubber, five- eighths of an inch thick. Holes-
were drilled into the edge of the rubber opposite the mercury
cups and copper wires inserted, which were connected with
the battery and induction coils in the manner which is*
schematized in Figure III. Beginning at the left hand end
(marked "L") of the hard rubber strip, the first pair of
opposite cups and each alternate pair along the strip were
connected with a coil of wire on one side, and with the battery
on the other. For purposes to be described later, were five
coils of wire which might be connected with these mercury
cups. The coil and the battery were connected, thus completing
the primary circuit. The other pairs of opposite cups which
alternated with these were all connected together on the one
side with an induction coil, and on the other with the tele-
phone. The induction coil and the telephone were joined,
thus completing the secondary circuit. The ten keys corre-
sponded to the ten pairs of mercury cups. When the first
key at the left hand, and each alternate key thereafter,
was pressed down by the arms on the drum-shaft so that
the platinum points dipped into the mercury, it would
close the primary circuit, for these keys joined the opposite
mercury cups which were connected with the battery. When
the second key at the left hand, and each alternate key there-
after was pressed down, it would close the induction cir-
cuit. Key 1 at the left hand end of the strip of hard rubber
matched the first pair of opposite cups of mercury and was
paired with key 2, which matched the second pair of the
opposite cups of mercury. These first two keys were operated
by the first arm at the left hand end of the drum- shaft. The
other four pairs of keys were operated by the other four arma
on the drum-shaft. Let us consider now only the first pair
of keys and the first arm at the left. As the shaft revolves,
the point of the arm which was in advance of the other was-
made to come in contact with the long end of key 1, and
pressed it down. After coming in contact with key 1,
the point of the arm could move through an arc of ten
degrees, keeping the platinum points in the mercury, and
thus closing the primary circuit, before the second point of
the same arm would come in contact with key 2. When
the keys were pressed down sufficiently to make the cir-
cuit, the points of the arm were made to slide by the ends of
the keys in such a way that the k§y was not released
until the arm had moved through an arc of twenty degrees.
A further revolution of ten degrees by the arm would press
key 2 down sufficiently to close the secondary circuit. If r
now, the arm continues to revolve, key 1 would be released
and rise up, breaking the primary circuit, but key 2 would
BHYTHM. 183
remain down while the arm moved through an arc of ten
degrees, keeping the secondary circuit closed for a time after
the primary circuit was broken. This would give a sound in
the telephone. The same process would be repeated with
each of the five pairs of keys and their corresponding arms.
If, now, the arms were set at an equal number of degrees apart
and the drum- shaft were made to revolve at a uniform rate,
the clicks in the telephone would be separated by equal inter-
vals of time, and not varying in intensity, pitch or quality,
these clicks would form the required series of auditory
impressions. If a change in intensity is desired, as it was,
the five wires connecting the different pairs of mercury cups
might each be connected with the five different coils which
were referred to above. These were set at different distances
from the induction coil (see Figure III.). As the different
primary coils were of the same size, the strength of the
induced current, and therefore the intensity of the sound,
would depend upon the distance at which the primary coils
were placed from the induction coil. They were placed at
just sufficient distance apart to make the sounds easily dis-
tinguishable from one another in a graded series of intensities.
By means of the key-board (marked a K") it was possible to
connect all the five wires in any way that was desired with
the five primary coils. The clicks might all be of the same
intensity, all different, or of two, three or four different inten-
sities. Whatever the variation, according to this arrange-
ment it would recur every fifth click. When variations every
fourth or third were desired, three or four arms were set upon
the drum- shaft and only three or four pairs of keys operated.
If the arms were separated by an equal number of degrees,
the series of clicks would still be regular. Two kinds of
arms were employed, those with a single end and those with
a double end. (Figure II. represents the double ended
arm.) By using both single and double ended arms on the
shaft, and operating the five pairs of keys, it was possible to
get an arrangement by which variations in intensity might
occur every sixth or eighth click. Taking all the possible
arrangements together, the operator might introduce a more
intense click every two, three, four, five, six or eight clicks.
Again, he might make a series of clicks which were composed
of two, three, four or five different intensities of sound.
By making the number of degrees between the arms on the
drum- shaft different, a difference in time-interval between
the clicks was produced. In the same way as with the dif-
ferent intensities, a longer interval of time might be made to
recur every two, three, four, five, six or eight clicks.
184 BOLTON :
The rate at which the drum- shaft revolved determined the
rate of the clicks in the telephone. This was controlled by
the fan regulator upon the chronograph. Faster or slower
rates were obtained by using smaller or larger fans. The
rate was determined by counting the clicks in the telephone
by a stop-watch. Bates between one click in two seconds
and ten in one second were possible. As the rate was a very
important factor, it will be given in all cases in the presenta-
tion of results. The " time" will indicate the interval be-
tween two clicks. The battery used consisted of 36 cells of
the Watson's patent.
A further method of testing the accuracy of the setting of
the arms upon the drum- shaft, which was done with a pro-
tractor, was to connect a time-marker in the primary circuit
and take the record upon a drum along with a tunning-fork.
It was found that setting might be accurate, but the drum-
shaft might vary between one and two hundredths seconds in
six seconds.
There is one particular in which an improvement might
have been made in the apparatus. It was this : When the
primary circuit was made, though the secondary circuit was
open, a faint sound was heard in the telephone with close
attention. The induction coil acts as an electric condenser, and
the telephone being extremely sensitive, betrayed the presence
of a weak current. This might have been avoided by making
a break in both wires leading to the telephone, in such a way
that the telephone would be wholly disconnected from the
induction coil, when the primary circuit was made. During
the entire experiment, only a single subject detected the
presence of this sound, and for that reason it may be disre-
garded. The telephone was placed in a different room from
the chronograph, where there was as little disturbance from
other noises as possible, especially from any noises that were
in the least suggestive of a rhythm.
When the experiment first began, the apparatus was set so
that about three or four clicks to the second were heard in
the telephone. The subjects were not informed in any par-
ticular in regard to the experiment. They were invited to be
seated and listen to the telephone. This they did, taking
very generally a rather critical attitude. They were then
invited to say anything that suggested itself to them, what-
ever the character. These statements were all carefully
recorded, and will be given in substance. The sounds sug-
gested most generally and immediately the clock. Other
suggestions were: slowly dripping water, galloping horse,
pile-driver, etc. After the subjects had been seated for a time,
during which it was apparent they were making a critical
RHYTHM. 185
study of the nature of the sounds, the statement most gener-
ally given, and voluntarily, was that the sounds were all
alike, and seemed to be separated by the same interval of
time. After this statement the subject paused, as if most
that could be said had been said. In some cases they asked
for particulars in regard to what they should look for. Some-
times, however, they went on to say that there was an
apparent change of intensity in the sounds ; the clicks seem
to group themselves by twos or fours, as the case might be ;
generally, however, it required some kind of a suggestion to
direct the attention of the subject to the grouping of the
sounds. An indirect method was preferred to a direct one.
In cases where the subject had spoken of the clicks seeming
like the clock ticks, they were asked if there was the same
difference of intensity or quality in the sounds as was
apparent in the clock ticks. This suggestion was sufficient
in many cases. The subject directed his attention then to
the matter, and if there was any tendency to make groups of
the clicks, it was apparent in a few moments. Sometimes it
was remarked that they had noticed such a grouping, but
had regarded it as a freak of their imagination, and did not
think it worth mentioning. Another method of directing the
attention of the subject to the grouping was to make a refer-
ence to the fact that they had said the sounds were all alike,
and then to ask why they had said sounds and not sound ;
did they suppose there was more than one sound 1 In this
case also, they replied frequently that they imagined that
there was more than one sound, but did not think it worth
while to mention the fact. In some cases it was sufficient to
ask the subjects to count the clicks as they heard them, and
then to ask how they counted. The reply was that they
counted four or two, as the case might be, and then began
again. Again it was noticed that the subject was uncon-
sciously keeping time, with the foot tapping to every fourth
or every second click. Such a subject was asked why he
tapped every fourth or second click, and so his attention was
directed to a grouping that was going on unconsciously.
Such indirect methods were usually successful, but there were
several cases in which indirect suggestions of this sort failed
of their purpose. Direct methods of tapping a rhythm with
the fingers or counting did not suggest anything beyond the
clock tick to two subjects. These persons possessed no
appreciation of music at all ; they could not u carry a tune,"
and yet were able to recognize some of the common airs when
they were sung or whistled. The general statement of the
remarks and answers of each subject will be given as fully as
it seems necessary. They will be abridged as far as possible,
186 BOLTON :
but the special features in the answers of each subject will be
mentioned. The treatment of special phases of these results
will follow, and then will be taken up the result of special
investigations that were suggested during the first part of the
experiment.
Subject 1. Some musical talent and training.
Time, .23 sec. The first suggestion was a 4-group. Subject
could suggest groups of two, three, five and six, but when he made
no suggestion either by tapping or counting, he returned to a
4-group. The third in each 4-group was accented, 1 but it was pos-
sible in the later experiments to accent any member of the group..
In general the first in all forms of grouping was accented. The
3-group was unpleasant and the 5-group was very difficult to main-
tain. Time, 1.14 sec. The most natural form of grouping was by
two. It was possible to get a 4-group, but when the subject made
no suggestion of any other group, he returned to the 2-group.
Time, .167 sec. The 6- group was most easily suggested. It had the
appearance of being composed of two 3-groups. The subject showed
a tendency with this rate to group the 3 and 4 -groups into higher
groups. Eight-groups of threes and 4- groups of fours succeeded very
well. It was not so easy or natural to make higher groups of fours.
Time, 1. sec. This rate produced a drowsy feeling. The subject
was inclined to make each click stand as the accented click in a
3-group, supplying the unaccented sounds between the accented in
imagination. When the subject was tired he noticed a tendency to
change the grouping frequently from two to three, and vice versa.
The subject showed a strong tendency towards 4- grouping in pref-
erence to all other forms of grouping, and yet during one experi-
ment, when the time was .208 sec, he found a 3-group more pleas-
ant than either a 2-group or a 4-group. The rate was too fast for
easy grouping by two. When he counted objects he counted them
by fours. Time, .323 sec. The subject was disposed to make a
4-group, and, even when every third sound was made more intense
than the others, he persisted in saying that he grouped them by
fours, but that there was probably a longer interval in the series
which disturbed the smoothness of his 4- groups. When his atten-
tion was called later to the accented clicks, he made no further
mistakes of longer intervals for accented sounds.
Subject 2. Some musical talent and training. Accustomed to
introspective work.
Time, .323 sec. The subject grouped by twos, visualizing the
pendulum. 2 He could suggest groups of three and four easily, the
four being more difficult than the three. Time, .263 sec. He
grouped the clicks by four, but the 4- groups were divided into two
2-groups. Time, .208 sec. This rate yielded easily and naturally to
a double 3-grouping. When he first listened to the telephone after
lr rhis accent consisted of an apparent increase in intensity with a change in pitch and
quality.
2 Almost every subject either visualized the pendulum or spoke of the pendulum-swing
movement sometime during the experiments. In either case it was a form of grouping.
When the rates were slow, the subject visualized the clock pendulum and made one
click come near the completion of each half swing. The clicks were then grouped by
two and were called the clock tick. In my own case and in some others there was a
strong tendency to sway the body with the pendulum. This was called the pendulum-
swing movement by the different subjects. It was quite visible at times. By this pen-
dulum-swing movement groups of two, thre*», four, six or eight were frequently grouped
into 2-groups. The first group, then, in the 2-group was accented or more emphatic than
the other, and a distinct pause seemed to follow the second group.
RHYTHM. 187
either a change of rate or at the beginning of a new experiment,
the clicks, did not group themselves, but in a short time the tendency
to group increased until it required the greatest efforts to hear the
clicks as a uniform series. The subject was able to hear the
clicks as a uniform series, only by imagining some one pound-
ing in the distance. It required a mental picture of some object-
ive thing that was perfectly uniform. When he gave himself up and
listened to the series as a whole, he fell into some kind of grouping,
which might or might not continue for any length of time. He had
a strong tendency to shift from one grouping to another. He com-
pared it to the optical illusion of the "stairs." The double 2-group is
confounded sometimes with the double 3- group. Time, .187 sec.
The subject said he got a compound 2 and 3- group, which by actual
count of the accents to which he tapped with the fingers, showed
he was making a double 4-group. 1 This subject was strongly dis-
posed to double groups of all sorts. Time, .323 sec. At this rate the
2-group was most naturally accompanied by the mental image of
the clock. Time, .263 sec. This yielded most easily to a 4-group,
which took the form of two groups of twos. Time, .167 sec. This
rate yielded at first to a 6-grouping, which was divided into two
groups of threes, but it did not persist there; he returned to a double
2-group.
The pulse seemed at times to impose a grouping in which the
clicks coming nearest in time to the heart-beat was accented.
When the subject gave his attention to breathing, it more generally
conformed itself to some grouping that was already going on.
Inhalation lasted during a 4-group and exhalation during a 3- group.
Subject 3. Considerable musical talent and training.
Time, .5 sec. The subject's first suggestion was of a 2-group, but
he immediately decided that a 4-group was more natural. He was
able to count almost any rhythm at this rate as far as twelve, and
the clicks seem to group themselves with the count. At first the
groups were apparently separated by a longer interval, which the
subject believed in the first place to be real. He was disposed to
regard the 4-group as the most satisfactory. Any grouping was
plainer when he counted. Diaphragmatic movements also accompany
the grouping. With indifferent attention there was no grouping.
The 4-group usually contained two accented clicks, either the first
and the third or the second and the fourth. The former were pre-
ferred. This rate was found to be most pleasing. It was animating.
The 5- group was difficult to get. A slight pause occurred between
the groups in every form of grouping. In the presence of the
chronograph, which gave a 6-rhythm which was composed of two
3-groups, the subject still grouped by four for a time, but this
tendency was finally overcome and the series yielded to the sugges-
tion of the chronograph.
When every fourth was accented, the subject being unaware of
this accent, said that the 4-group only was possible, for there
appeared to be a longer interval between every four clicks which
made any other grouping impossible. When the accent was
strengthened, he said the interval had been lengthened. This long
interval might come anywhere within the group of four, but it more
generally came between the groups. When two stronger clicks
followed by two weaker ones formed the series, the subject said the
*It is not unfrequent for a subject to mistake the actual grouping which he is making.
Sometimes a subseet is so disposed to a particular number that he persists in saying
that he gets groups of that number, when it is perfectly evident a greater or less number
of clicks according to the circumstances is grouped with the accented clicks to which
he taps.
188 BOLTON :
rate was slower. He grouped the series by fours, but it appeared
as though two long sounds followed by two short ones formed the
group. When three strong sounds and one weak one formed the
series, he still grouped by four. The first two in each group seemed
to be of the same length, the third was longer and the fourth very
short. During all the experiments the subject confounded stronger
clicks with long intervals, and was never able to tell the difference
between a strong sound and a long interval. He was surprised
when told afterwards that the longer interval had been caused by
accenting one sound. Time, 2.304 sec. The subject visualized the
pendulum, but said the pendulum seemed to reach its full swing
before the click corresponding to the swing was heard. The clicks
seemed to delay too long. Time, .323 sec. Every third sound was
accented. The subject had a strong mental habit for grouping by
fours and was greatly puzzled by this accent on every third, which
he said was a longer interval and broke up his tendency to form
groups of fours. Time, .208 sec. Every third was accented. The
subject forms 6-groups, which were accented upon the first and
fourth and a long interval appeared between the groups of six.
Time, .137 sec. When the series was composed of clicks of three
different intensities repeating themselves in the same order, the
3-groups were again grouped by four generally, though the subject
eould suggest groups of three 3-groups.
Subject 4. Some musical talent.
Time, .288 sec. The clicks suggested the clock-tick. The subject
could group them by twos, but he found it more natural to groap by
fours. It has long been a mental habit with him to make groups of
four of any objects or impressions that would admit of any kind of
grouping. He counts by four and groups- the puffs of a locomotive
by four. Four objects or impressions of any sort standing together
have always arrested his attention. He found it possible to group
these clicks by two, three or five when he made a suggestion either
by counting or tapping with the fingers, but when the suggestion
was stopped he returned to a 4- group. In every kind of grouping
the first sound was always accented. Time, ,115 sec. The subject
said the grouping was by four and was requested to tap the
accented click in every group. In six trials for five seconds each,
he tapped just five times during each trial, showing he made a group
of four in one second. The actual number of clicks to the second
being 8.6, it was apparent that he was making a much larger group
than four, probably an 8 -group. When asked to make a 3 -group
and tap the accented click in each group, the results were nineteen
taps in fifteen seconds, showing that his groups were not far from
six instead of three clicks to the group. 1
The 3- group was really a 4- group in many cases. Between each
group of three occurred one click, of which no account was taken.
It seemed to him something like this when he counted : 1, 2, 3, 1 —
1, 2, 3, 1-1, 2, 3, 1.
When the subject gave attention to the pulse, the number of
clicks coming between the beats of the heart formed a group. The
click which came nearest in time to the heart-beat seemed always
to correspond to it. The breathing adjusts itself to the 4-rhythm.
Inhalation lasts during one group of four, and exhalation during
another. In this way the 4-groups were grouped by two. Time,
.156 sec. By forming a mental image of the pendulum, or of some
object moving up and down, he was able to make double 4-groups,
*It is probable that the primary grouping: was two, and these groups of two were then
united into larger groups of three and four.
RHYTHM. 189
which corresponded to the full swing of the pendulum. These
three rates were given in rapid succession. Time, .536 sec. The
4-group was very clear. Time, .268 sec. The 4-group was un-
pleasant at first, but he gradually became accustomed to it. At
first the rates seemed too fast. Time, .536 sec. With this latter rate
the 4-group seemed to divide into two groups of two during the
second trial. Time 1.072 sec. The 2-group seemed most natural
and the subject felt a strong tendency to form higher groups of
twos. After the subject became accustomed to this rate, he was
more inclined to form 4-groups than 2-groups, but still the third
and fourth clicks seemed at times more like a 2-group than a part
of a 4-group. The general effect of this last rate was soothing.
Time, 1.66 sec. This rate was soporific ; it was possible to form a
2-group, but he did not do so spontaneously. Time, .323 sec. The
subject grouped the clicks by four and found difficulty in suggest
ing a 3- group. When every third was accented, he persisted in
grouping by four. Again he was asked to suggest a 3 -grouping,
and he succeeded in doing so. The subject was unaware of the
accent, and expressed surprise that he could group by three, and
found it easier than grouping by four. The following rates were
given in rapid succession. Time, .268 sec. The 4-group was very
clear and pleasant. Time, .17 sec. The subject grouped by fours,
but felt a confused irritating feeling. There was something added
onto each group of four. Time, .134 sec. This rate recalled the
sound of a locomotive. He visualized a revolving wheel, during
each revolution of which he counted four. Time, .116 sec. He
still groups by four, but the 4-groups are grouped by two, a strong
and weak group together. When every eighth was accented, he
grouped by eight. There was a distinct pause along with the
accented sound. Time, .134 sec. Every eighth was accented. The
8-group divided into two 4-groups, with a pause after the second
group. The 4-groups were grouped by two with the pendulum
swing. Time, .268 sec. Every eighth was accented. The subject
grouped by four and associated the pendulum swing with the
groups of four.
Subject 5. Some musical talent.
Time, .288 sec. The subject was most naturally disposed to a
4-group, and found it difficult to get any other. Time, .78 sec. The
2-group was most natural with this rate. The 3- group was pleasant
and easy when it was suggested. Time, 1.44 sec. It was easy to
form 2-groups, but other groups were impossible. These three
rates were given in rapid succession. Time, .353 sec. The 4-group
was most natural. Time, .183 sec. The 8-group was most agree-
able with this rate. Time, .156 sec. The subject found the 8-group
most natural.
Time, .3 sec. When every fifth was accented, the subject made
groups of four, accenting the four, and said that there was a rest
between each group. He found it quite as easy also to accent the
first. When his attention was called to this pause between the
groups, he decided that the groups contained five clicks, in which
the fourth was accented. When the first and third clicks were
made more intense, the subject was greatly puzzled for a time, but
decided that the series was compounded of a 2-group and a 3-group.
Time, .969 sec. He was able to form 3 and 4-groups. The series
was associated strongly with the clock, and for that reason the
4-group tended strongly to divide into two 2-groups. Time, .323
sec. He found the 4-group most natural and pleasant, and when
190 BOLTON :
he attempted to form groups of three, they would immediately run
into fours. Time, .208 sec. The subject formed a long group of the
clicks which he thought was an 8 -group. The rate was too fast for
easy grouping by four, and his attention seemed to waver between
a 4- group and a longer one which he thought was an eight.
Subject 6. Some musical talent and training.
Time, .3 sec. When the subject first listened to the telephone he
found no tendency to form groups of these clicks. Even after
repeated suggestions the subject did not comprehend what was
desired or to be looked for. He was asked to tap an accompani-
ment to the sounds with four fingers. After a time he found
himself accenting the third, and grouping the sounds by four.
When he tried the suggestion with three fingers it did not succeed
very well. When two were tried, the subject decided that the
4-group was a combination of two 2-groups. In the same way he
was able to group the sounds by six, but the groups divided easily
into two 3- groups. On the whole the 2- group was the most
natural with this rate. Time, .156 sec. The 4-group prevailed and
easily combined into 8-groups. Time, .78 sec. The 2-group was
most natural, but the subject was able to suggest the 3-group easily.
Time, 1.44 sec. Even a 2-group was difficult to get. The time
seemed to be too long. Time, .353 sec. Groups of two, three and
four were all possible. The first click in all groups was accented,
and the third also in the 4 -groups. When he suggested a 6 -group, it
divided easily into two 3-groups or three 2-groups. The 8-group di-
vided readily into two 4-groups. It was difficult to get a 5-group. The
5-group tended to run into a 6-group, which then divided easily into
two 3-groups. Time, .288 sec. The 4-group was most natural, and
readily combined into double 4-groups. Time, .156 sec. The 6-group
was most natural. When the telephone was disconnected and the
chronograph continued to run, the grouping always began with the
first click in a new group. Time, .78 sec. The 2-group was the
most natural. Time, 1.66 sec. There was no spontaneous grouping
with this rate. It was too slow. The subject found it convenient
to regard the click as an accented click in a 4-group, supplying the
three intermediate sounds in imagination. Time, .323 sec. At this
rate, the subject showed strong tendency to muscular movements,
either to tap with the finger or toe upon the accented click, some-
times to nod the head or sway the body. He found this rate very
favorable for voluntary changes of the grouping, which he did
either by counting or tapping with the fingers. The general
emotional effect was depressing.
The pendulum-swing movement or 2-rhythm was an important
factor in all his groups. The 6-group was usually composed of two
3-groups, and the 8-group of two 4-groups. Time, .167 sec. The
4-group was very clear and pleasant, and the subject tended to
group them by the motions of the pendulum. When every third
click was strengthened, the subject grouped by threes, and made
the 3-groups follow the motions of the pendulum. In this way the
series produced an exciting effect. Time, .134 sec. The 4-group
was plain and distinctly grouped by twos by the pendulum- swing.
This rate was also exciting and animating. When every sixth click
was accented, the grouping lost its exciting effect. When the
series was composed of clicks of three intensities, the strongest
first, the clicks were grouped primarily by threes and these
3-groups were again grouped into 4-groups. When the time was
changed to .167 sec, and three grades of intensity retained, the
higher grouping of 3-groups by four ceased. Time, .137 sec. When
RHYTHM. 191
<every sixth click was accented, the series was grouped by six, and
the six groups were again grouped by the motions of the pendulum.
Time, .167 sec. Every sixth was accented. The grouping was still by
six, but the 6-groups did not group by two. With the slower rate
the 6-groups did not seem so compact as with the faster rate.
Time, .208 sec. Every sixth was accented. The 6-group was diffi-
cult to grasp. The time was too long and the group tended to
divide into two 3- groups. Time, .263 sec. Every sixth was
accented. There was greater difficulty still in grasping the 6-group,
on account of the tendency to divide the group into two 3- groups.
Time, .323 sec. Every sixth was accented. The subject now
grouped the series by four in spite of the accent upon every sixth.
When the subject heard the sound of the chronograph, which was
rhythmical, he grouped the clicks according to this rhythm, which
in this case was a double 3-rhythmical. Time, .3 sec. Every fiftn sound
was accented. The subject grouped by fours, but the accent came
in a different place in each group. It shifted one place further to
the right. When a 6-group was suggested to the subject, the
accent changed its position in the opposite way. It shifted its
position one place to the left in each group.
Subject 7. Some musical talent and training.
Time, .3 sec. Almost immediately the series divided into groups
of four, and soon after the 4- groups were grouped by two. With
the suggestion of tapping, the subject was able to group by two,
three or five. Time, .78 sec. The 2-group could be suggested
only with difficulty. The 3 and 4-groups were not at all pleasant.
Time, .156 sec. There was no distinct grouping. The series seemed
to rise and fall in intensity at regular intervals. At times he had a
"dreadful" feeling that the chronograph was slowing up and about to
stop. The subject had observed this grouping of sounds in the puffing
of a locomotive. He had not noticed a definite number in the group.
The sounds simply rise and fall in intensity. Speaking of the
8 -group, the subject said he had a feeling of not being able to
" round up " until he came to eight. It seemed natural to stop at
eight, and start over again. This group was accompanied by a
feeling of completeness. During inhalation, the clicks seemed to
come faster, and slower during exhalation. In all forms of group-
ing the subject felt a muscular sensation in the stomach and intes-
tines. He also felt a strong tendency to beat time with the thumb.
He had been taught to do so when quite young.
Four- grouping is a kind of mental habit with him. When the
series was accented upon every fifth, he still grouped by four,
making a pause between each group in which he pronounced the
word "and:" 1, 2, 3, 4, and 1, 2, 3, 4, and 1, 2, 3, 4. In the same
way the 3-group appeared as 1, 2, 3, 1,-1, 2, 3, 1,-1, 2, 3, 1. In the
4- group the first and third were always accented. In other groups
iihe first was accented. When the rate was .72 sec. or .156 sec. the
grouping did not come without suggestion by muscular movement
or counting; with intermediate rates the grouping by four was
wholly involuntary.
The subject could group 4-groups by two very easily, but it
required an effort to group them by four, or 8- groups by two. The
8 -group was generally composed of a more and a less emphatic group
of four.
Time, .268 sec. A double 4- group was the most natural and easy.
One group was more emphatic than the other. Time, .263 sec.
Though the time was only slightly changed, the subject thought the
previous double 4-group changed into a more perfect 8-group.
192 BOLTON :
There was not such a strong division into two 4- groups. Time, .208
sec. This rate gave a " better 8-group " than the previous one.
Time, .167 sec. The groups did not separate distinctly. There was
a kind of confused feeling about the clicks. Time, .137 sec. The
confused feeling with the previous rate was more apparent still.
When every sixth was accented, he grouped by six and the
6-groups were grouped by two with the pendulum- swing movement.
Subject 8.
Time, .3 sec. The subject took a critical attitude. He had no
preference for any grouping. He could count any number as far as
ten, and the series seemed to group itself according to the count.
In the longer groups, groups of two were frequent. Time, .156 sec.
The 8-group was the most suitable. It was composed of two
4-groups, and each 4-group of two 2-groups. Time, 1.44 sec. It was
not possible to form any grouping. The rate was too slow. The
subject has noticed rhythms in the sounds of a mill-wheel, locomo-
tives and fans. He was not aware of any definite grouping of the
sounds.
Subject 9. Considerable musical talent. Long and careful training
in music. Accustomed to introspective study.
Time, .3 sec. The subject adopted a critical attitude and gave his
attention to the nature of the sounds. At first he was inclined to
believe that they were all alike in intensity, but then he thought
every third was stronger than the rest. For a time the interval
between the clicks seemed to be irregular, but he soon discovered
that this irregular interval might occur anywhere he chose to put
it. In a short time his tendency to find groupings of the clicks
grew so strong that it required an effort to hear the series uniform.
Such an effort was akin to the feeling of * 'looking long into the
future." The grouping tendency had to be restrained. Time, .3
sec. The 4-group was so plain that he did not discover the fact
that it was imaginary and was completely surprised that the illusion
was so complete. It was then more than ever an effort to hear a
uniform series of single impressions. He said, "I find no rhythm as
long as I hold my breath and stick to it." "I get hold of one click
to compare it with the succeeding clicks, but I can't hold onto more
than eight or nine." The simple suggestion of any grouping was
sufficient to produce that grouping. Groups of two, three, five, six
and eight follow immediately the suggestion of any of them. A
group of seven was more difficult. Groups of all numbers were
generally accented upon the first, but the accent could be volun-
tarily changed.. In the 8-group the subject had a tendency to accent
every other one. The grouping was generally accompanied by
visible motions of the head and lips. A slight feeling of muscle
tension in the ear and back of the scalp marked one group from
another. There was a feeling of innervation of the muscles con-
nected with attention.
When the attention was directed to respiration the grouping was
not affected. Respiration was more inclined to follow the group-
ing. The heart-beat coming in about the same time as the accent
in a 4-group, tended to coincide with it. Time, .2 sec. The tendency
to group was still present in a small degree. The sound was
quieting. It suggested slowly dripping water. Time, 1.5 sec. The
suggestion was of a big clock. After listening to a fast rate for a
time and then to the rate of .687 sec, he decided that he felt no
tendency to group the sounds of the latter. Time, .116 sec. The
most natural group was eight, with a slight tendency to divide into
RHYTHM. 193
two 4-groups. Time, .134 sec. The subject found a 4- group more
natural than an 8 with this rate ? but felt some tendency to make a
double 4-group instead of a simple 4-group. Time, .116 sec. An
8-group composed of two 4-groups was most pleasant. Time, .268
sec. His most pleasant group was two, but these groups tended to
combine to form double 2 -groups. When two strong clicks followed
by two weak ones formed the series, he grouped by fours, but
accented the second and fourth. He described the phenomonon as
a summation in the second : "The after-image of the first was left
to the second to increase its strength." When the subject heard
the rhythm of the chronograph, he grouped the sounds accordingly.
Subject 10. Some musical talent and training.
Time, .3 sec. The 4-group appeared immediately. The subject
could suggest other groups of three, five, six and eight. The 3-
group was accented upon the first, and the 4-group upon the first and
third. During one experiment the subject said the accents in the
4-group were not distinguishable, but the groups were separated by
a slight interval. The 5-group was accented upon the first and
fourth, the 6- group upon every other one, and the 8-group was a
repetition of two 4-groups. Higher groups of threes as far as four
were easily obtained. The first and third groups of threes were
accented. Higher groups of fours were not easy or distinct.
During all the experiments unconscious movements in the tongue
were present. A slight muscular contraction took place with the
accented click. Other movements of the head, trunk, feet
and hands were visible, and the subject found it difficult to restrain
them. Time, .134 sec. These sounds were grouped by eight and
the grouping was pleasant and animating. Time, .156 sec. The
first suggestion was of a 6-group. The 8-group was difficult. Time,
.268 sec. This yielded to a 4-group, which the subject said required
about the same time as the previous 8-group. Time, .78 sec. The
2- group was the only one practicable. The suspense for others was
too great. Time, .116 sec. During this experiment the rate yielded
most easily to a double 4-group, and when the time was changed to
.45 sec. he had a similar feeling with the group of two, but one click
stood in the place of the 4-group with the previous rate.
Time, .116 sec. Every eighth was accented. The subject got a
very pleasant and "harmonious" 8-group. Time, .134 sec. Every
eighth was accented. The subject was less animated. He said,
"The group was more staid and steady. It had lost its tones."
Time, .17 sec. Every eighth was accented. It now required an
effort of attention to get the 8-group. It grew more pleasant as
the subject became accustomed to it. Time, .268 sec. Every eighth
was accented. The grouping was by four. Sometimes the subject
accented every other one and felt disposed to count thus : one and
two and three and four, and repeating this between the accented
clicks.
Subject ll. 1 Some musical talent and training.
Time, .5 sec. The 4-group suggested itself immediately. The
first and third clicks were accented, the first stronger than the
third. Sometimes the third might be stronger than the first. It
was possible to accent the second and fourth. When the subject
gave close and critical attention to the sound, there was no tendency
to grouping. The grouping seemed most clear with an indifferent
state of mind. He showed a decided preference for 2 and 4-groups.
Time, .25 sec. The 4-group was most natural. Time, .115 sec. This
*The subject knew beforehand that this was to be an experiment in the rhythmical
grouping of sounds.
194 BOLTON :
rate yielded to an 8- grouping, each group being composed of four
strong and four weak sounds. At other times with this rate the
sounds seemed to rise and fall at regular intervals, which the subject
described as a waxing and waning of the attention. Time, .167
sec. The subject grouped by four, but felt a straining for a larger
group. Time, 1.67 sec. He grouped by two and visualized the
pendulum. One click came during each half swing. When the
subject gave attention to his breathing, he made an inspiration last
during the time of one click, and expiration during the time of
another. The first click was louder than the last. Time, .115 sec.
When the subject gave attention to his pulse the groups corre-
sponded to the time of the heart-beats. The click which came near
the beat was louder and became the first in the group. The pulse
seemed to reinforce the sensation of the sound. When the attention
was directed to respiration, the clicks increased in intensity during
inspiration and were grouped by two and decreased in intensity
during expiration. He visualized a curved line which rose during
inspiration and fell during expiration. Smaller undulations in the
larger curve corresponded to the 2-group. A melody always
appears to him as a zigzag line, in which the rises correspond with
every two notes. Time, .156 sec. He grouped the clicks by eight
and visualized an ellipse with four points upon either side. The
clicks seemed to locate themselves on these points.
The subject showed a strong tendency to muscular movements.
He felt an impulse to dance, clap the hands and tap the toes and
fingers upon the accented click. When the rate was .286 sec, this
tendency to muscular movements was stronger than with the other
rates. There was something animating about this rate.
Time, .3 sec. Every fifth was accented. The clicks were
grouped by five. The accented click always appeared as the
fourth in the 5 -group and longer than the others. When this
click was further increased in intensity, it seemed very much
longer than the rest and appeared as an extraneous sound
which did not enter into the group. The other four sounds then
formed a group by themselves. When every sixth was accented,
the accented sound again appeared as an extraneous sound. It
simply disturbed his mental habits of forming some other groups.
When two clicks in every five were made stronger with one weak
click between the two strong ones, the grouping was still by five
but it was a combination of a 2-group and a 3- group. When three
strong and two weak clicks formed a group, it was composed then
of a 3-group and a 2-group. The 3-group contained two strong
sounds and one weak, and the 2-group one strong and one weak.
A short pause came after the fourth sound, which made it impossible
to make the 5-group appear as composed of a 2-group and a 3-group.
Time, .268 sec. Every third sound was accented. This accent
simply broke up the tendency of the subject to group by four and
did not compel him to group by three. When every sixth was
accented, he grouped by six, and accented the first and fifth, but
there was a strain towards a 4- group. Time, .167 sec. Every sixth
was accented. With this rate the 6 -group was pleasant and did
not tend so strongly towards a 4-group. Time, .137 sec. Every
sixth was accented. The 6- group was pleasant, and it tended
to unite into higher groups of two with the pendulum- swing move-
ment.
Time, .208 sec. When the subject listened to the sound of the
chronograph, which made a distinct and strong 8 -rhythm, he was
unable to form any other group than eight. The 8- group was com-
posed of two 4- groups, the first of which was much stronger than
EHYTHM. 195
the second. When he listened to the chronograph, which gave a
6-rhythm, which was composed of two 3-rhythms, he was unable
for a time to get anything but a 6 -group, but this faded out with
continued effort and gave place to his previous 4-groups. The
4- groups were then grouped by two with the swing of the pendulum.
Time, .134 sec. Every eighth was accented. The grouping was by
eight, and the 8-groups were then grouped by two.
Time, .116 sec. Every eighth was accented, and the grouping
was by eight, and pleasant. Time, .134 sec. Every eighth was
accented. The subject took no spontaneous interest in the 8- group
at this rate. The period seemed to be too long. "It breaks off
with a dead end," he said. Time, .17 sec. Every eighth was
accented, but the grouping was by fours. The accented click was
simply a disturbing element. The series did not group easily by
either four or eight. Time, .208 sec. Every eighth was accented.
The grouping at this rate was distinctly by four. The accented
click acted somewhat as a disturbing element. When every fourth
was accented at this rate, the 4- grouping became pleasant, and the
accented sound was the first in each group. The 4-groups were
grouped by two with the swing of the pendulum. Time, .17 sec.
Every fourth was accented, but the time seemed to be too fast for
a pleasant 4-group.
When every sixth was accented, and the time .323 sec, the
grouping was by three, but the tendency to a 4 -grouping was so
strong that it was possible to get a 4-group in which every sixth
sound was accented, the accented sound shifting its position in the
group. The accented click seemed longer, and a longer interval
followed it. When a very weak sound was followed by a very
intense one, the sound of the loud click spread itself over the
weaker one.
Subject 12. Considerable musical talent and great interest in
music. Accustomed to introspective study.
Time, .3 sec. The subject began immediately to count the clicks,
accenting every third. He unconsciously rocked himself in the
chair to keep time. He thought the rate slowed up at times and
then quickened again. The grouping was changed from three to
four by simply thinking of the number. He believed there was
some unconscious muscular movement about the change from one
rate to another. He could suggest a change by simply tapping with
his fingers. When he changed from a 3-group to a 4-group, the
4-group seemed too long at first, though he became accustomed to
it. In a short time the grouping seemed to change of itself into
three and then again into four. The 4-group was inclined to fall
into two 2-groups, the subject unconsciously nodding his head to
every other sound. He was able to suggest a 5 -group, in which
the first and third were accented, the third more strongly.
He could accent any click in the group, but the first and the third
seemed easiest. Time, .156 sec. The 6-group appeared imme-
diately and spontaneously, and then broke up into two 3- groups.
He suggested a double 4-group ; which gave rise to a feeling of a
slower pace. It was not so distinct as the double 3-group. This
had a kind of impelling force. The subject attempted to step in
time with the double 3-group ; and then with the 4-group. The
double 3-group required a sprightly step. It was exciting. The
4-group at this rate did not appeal to him; it didn't take hold. This
rate was more stimulating than the previous one. Time, .115 sec.
The subject dropped into a 4-group, but the three was found more
stimulating. It was difficult, however, for him to put aside the
196 BOLTON :
previous rate, and adapt himself to the new one. One click in
each group, nowever, seemed distinctly louder than the rest.
When he grouped by four, it easily passed into an 8-group, but the
8-group was not so clear as the 6- group! He imagined a wheel
going around, making six clicks to a revolution. When he changed
the telephone from one ear to the other, the grouping changed
from six to a double 4-group,and persisted for a time. The 5-group
came only with difficulty. Time, .76 sec. The grouping was by
twos. The subject imagined the clock at home. The 3- group was
suggested by an image of a musical conductor beating time. Time,
1.44 sec. The subject gets the rhythm of the pendulum swing with-
out suggestion. He suggests also a 3-group, which recalls the time
of church music. Time, 1.66 sec. He finds it easy to imagine inter-
mediate sounds between the actual clicks, and these he groups by
three, the real click being the accented click in the 3-group. Time,
.286 sec. In order to obtain a notion of a rhythmic series— one of
uniform intensity — the subject turned his attention backward, and
saw a series of images to which he was adding one all the time. He
throws his attention upon what comes, and studies the nature of
the noise to see if the timbre is the same. It is a comparative
effort. But in spite of all efforts the series groups into a 2-group
at times. When a relay sounder was connected in the circuit of
a vibrator, which made 20 vibratious to the second, the subject
was still able to effect a grouping of the sounds into either 3 or
4-groups by tapping with the fingers upon the table. When he
dispensed with the suggestion, the clicks of the relay signal were
perfectly uniform, except perhaps a slight waxing and waning in
intensity, due probably to the waxing and waning of the attention
towards the sound. There was no real grouping.
When a longer interval was introduced every fourth, the clicks
came in a group of four, but there was nothing satisfactory about
the group. The clicks did not form an organic group. Each group
of four stood rather as a single compound impression. There
was no organic relation between the separate clicks in the group.
When the rate was rapid, the groups of four were grouped into
higher groups, the groups of four standing as single impressions.
When the rate was slow the long interval might come between the
groups or anywhere within it. There was something wanting,
something to be looked for in the interval.
As the nature of the group, the subject described his feeling as a
tendency to go back when he had heard three or four clicks, as the
ease might be. He says he has a " mouthful "—a unity— and when
he has one. he seeks to get another. The same process continues
to repeat itself. When he directed his attention to the timbre of
the click, he got no grouping, but when he looked at the series as a
whole, the grouping was clear and spontaneous. There was not,
however, necessarilv an accent in the group.
Subject 13. Considerable musical talent and training. A lover
of 2-4 music.
Time, .285 sec. It suggests the gallop of a horse— a short gallop
—and the clock. There is a breathless feeling about it. It is the
sound of car wheels— the whole train. It has a double vibration.
The clicks are grouped by two or by four. The group seems to
close with a rising inflection ; the last is apparently accented
lightly, as the first is strongly. The 2-group prevailed over the
four. Parts of "Erl King" are suggested by this grouping. An
objective suggestion was displeasing to the subject. The subject
preferred a mental suggestion in order to change the grouping from
two to anything that was desired. By such a suggestion the sub-
RHYTHM. 197
ject was able to get most any group up to eight. The eight group
was not clear ; the accents were not sufficiently prominent. The
shorter measures are more strongly accented. Time, .115 sec.
This rate had a bad effect ; it was tormenting. The grouping was
by four of a particular pitch, followed by four of a lower pitch.
The subject might group the clicks by two in the same way. but
with less clearness. Time, .352 sec. This rate suggested something
going around, and every other sound was accented. When the
3- group was suggested, the first click was accented, and the group
closed with a rising inflection. Higher groups of 3- groups could be
obtained as far as four. The groups seemed to rise and fall in inten-
sity. At this rate also the short groups were more strongly
accented than the long. When the subject suggested a 4- group,
the first and the third were accented, the first probably stronger.
The 4- groups may be grouped again by four. Twenty was the
greatest number of clicks that the subject could grasp easily in
this way. The grouping becomes lost and disconnected with larger
numbers. The first groups in the larger groups were of greater
intensity, and the last of a lesser intensity. The intensity of each
succeeding group seemed to be less. This rate was said to have the
most " aesthetic effect." Time, .268 sec. The 2-group was most
easy ; a double 2-group was pleasant. The general effect of this
rate was a hurried feeling. The previous rate had been restful.
Time, .156 sec. The 4- group was most natural, and was accented upon
the first and the third. The 6- group appeared without volun-
tary effort. There may have been a mental suggestion of the six.
Time, .78 sec. There was no real grouping. It seemed painfully
slow. Time, 1.44 sec. The subject supplied two intermediate
sounds between the clicks, and grouped by three. The actual click
was the accented sound in each group, and came first. Time, 1.66
sec. The subject supplied three intermediate sounds between the
clicks, and grouped by fours. The real sound came first. Time, .134
sec. The double 4-group was most natural, and the subject breathed
with it. When every eighth was accented, the subject did not
become aware of the accent. The grouping was spoken of as being so
strong that it could not be gotten rid of. The groups of eight were
grouped by two with the swing of the pendulum. The clicks in the
8-group seem to decrease in intensity from the beginning to the
end. Time, .116 sec. Every eighth was accented. The movement
was the same as with the previous rate, or perhaps in place of the
pendulum movement the subject visualized an object moving up
and down, the upward motion lasting during the time of an
8-group, and the downward motion during another 8-group.
There was apparently a longer pause after the second group. The
subject felt a strong tendency to nod the head, and keep the time
by tapping the toe. Time, .17 sec. Every eighth was accented.
The 8-group lacked completeness. It was not so smooth as
the 8-group before ; it was distinctly divided into two 4- groups.
The accented sounds were generally unpleasant. The subject " has
not the restful impression of evenness" which had characterized
the uniform series.
Time, .323 sec. When the clicks were all of the same intensity,
the slightest suggestion of any sort was sufficient to cause the clicks
to fall into the group suggested. Even when the attention of the
subject was not called to a suggestion, and the subject apparently
did not attend, it would change the grouping to that suggested.
At times the subject had a feeling which was described as
" awful," that the chronograph was slowing up and about to stop.
When stronger clicks were introduced, the effect was unpleasant.
198 BOLTON :
The following rates were given in rapid succession: Time, .268 sec.
The clicks were grouped by two, and the 2-groups seemed to rise
and fall in intensity at regular intervals. Subject could suggest
other groupings, but it drifted back to this, unless the subject kept
up the suggestion of some other. Time, .208 sec. The grouping
was by four. The rate was unpleasantly fast for a time. Time,
.134 sec. The grouping was by four ; the 4-groups seemed to rise
and fall in intensity, every other one being more intense. The
subject unconsciously breathed with this secondary grouping.
Every eighth was made more intense. The subject did not detect
the accent, but said the grouping by eight was so clear that it could
not be avoided. The 8-groups tended to group into 2-groups.
Time, .116 sec. Every eighth was accented. The clicks were
grouped by eight, and the 8-groups were grouped by a wave -like
motion. There appeared to be a longer interval between every
two groups. Time, .17 sec. Every eighth was accented. The
grouping was primarily by two, and the 2-groups were grouped by
four. The intensity of the clicks seemed to decrease from the
beginning to the end. The grouping was rough in comparison with
that for the previous rate. This form of grouping gave place finally
to a double 4- grouping, and the subject was strongly inclined to
keep the time by nodding or tapping with the toe. Especially
strong was this impulse when strength of the accent was increased.
Time, .208 sec. Every eighth was accented. The 8-group was now
more distinctly divided into two 4-groups. This grouping had
more " dignity and force, but was not so tripping as the fast rate
was." The 8-group was not so complete as it was with the faster
rates.
Subject 14. Some musical talent and training.
The first suggestion of a grouping was by eight, and the 8-group
was divided into two 4-groups. When a 2 -group was suggested
the subject agreed that he could get it, but the 2-groups were again
grouped by two into 4-groups, and. the 4-groups by two into
8-groups. A 6-group was suggested by counting six, but there
seemed to be a division corresponding to 4-groups. The subject
was under the impression for a time that there was a longer interval
or four different intensities of sounds which made this 4- grouping.
The 4- group was accented upon the first and the third. The
3- group did not succeed very well. The subject seemed to have a
habit of forming groups of two, and the strongest kind of a sugges-
tion was not sufficient to put it aside for a 3 -group. Time, .156 sec.
The 8-group, which was divided into two 4-groups, was the most
natural, ana seemed to prevail over all others. Time, .78 sec.
The 2- group was most easily obtained, but it was possible to sug-
gest either a 3-group or a 4-group. The subject was not sure
whether he preferred a 2-group to a 4-group. He also found the
3-group quite pleasant. Time, 1.44 sec. The 2-group was most
natural, and the subject could still suggest either a 3 or a 4-group,
but when he dispensed with suggestion, he returned to the 2-group.
The subject has noticed rhythms in the sound of mill wheels.
When he gave his attention to these sounds he visualized a series
of points on a line which he counted by four or two. When he was
asked to count a series of dots, he said they were divided off into
twos by a bracket above them. It has always been a habit with
him to count objects by two.
When every fifth was accented, he grouped by five; the accented
click came fourth in the group, and it seemed longer than the rest.
When the accented click was made more intense still, its time
seemed longer than the rest. When one of the five was made
RHYTHM. 199
weaker than the rest, they formed a somewhat irregular group that
was unpleasant. The weak sound caused a disturbance in the group
which was not present when a louder sound was introduced.
When all the clicks were made more intense, the rate seemed to be
slower than at other times.
Subject 15. Some musical talent and training.
Time, .3 sec. The sound suggested the clock. It was more easy
and natural to regard every other one stronger. Groups of three,
four and five were suggested. The 4- group was the most natural;
the first and third clicks were accented. At times the 4-group
seemed to divide into two 2- groups. When the subject attempted to
compare the 3-group with the 4-group in point of their agreeable-
ness, the three group appeared as three, with one sound coming
between the groups, thus : 1, 2, 3, 1, — 1, 2, 3, 1,-1, 2, 3, 1. This
extra sound seemed to occupy a blank space between the three
groups. During other experiments afterwards, the 3-group
appeared in this form. The 4-groups were easily grouped by two.
They would combine into no higher groups as simple 4-groups.
The subject was able with great effort to combine two double
4-groups. When the subject counted objects, he usually grouped
them by twos. The objects seemed to be joined together by bars.
Time, .57 sec. This rate was very quieting. The 4-group was most
natural. The first and third or the second and fourth might be
accented. A longer interval appeared between the separate groups.
A long interval generally follows the accented click whether it is
imaginary or real. The subject regards real accents as extraneous
intruders. They introduce a long interval, and for that reason the
series seems irregular. By irregularity he understood a difference
in time interval of the clicks. The accented click seemed nearer
to the preceding click than the others. When two real accents of
unequal intensity were put into a group of eight, the interval fol-
lowing the more intense click was the longer, and gave to a series
a very irregular appearance. When the accented clicks were
dropped out, the series became regular again.
Time, .268 sec. This rate was very favorable for voluntary
changes of grouping. He could suggest any grouping that he might
desire within limits. During every experiment the subject mani-
fested a strong tendency to some kind of muscular movements.
Any kind of muscular contractions would suffice as a suggestion of
a grouping. He said he either counted the clicks or made the
proper muscular adjustments for counting. There was mental
counting always at the start. He made unconscious movements
with the eyelids. Motions of the head were clearly visible the
whole time. When the subject was asked to restrain all movements
of which he was conscious, he said there was great difficulty in
keeping the grouping. The telephone was disconnected, and the
subject was requested to restrain his muscular movements or
attempts to count. When the telephone was connected again, he said
that the grouping had kept up during the interval. Although he
had restrained all visible motions, slight muscular contractions
were observed in the eyelids at the proper intervals of the accented
clicks. He said it was possible to keep the grouping by imagining
a series of colors passing before the eyes. He spoke of a feeling in
the eyes as " muscular color sensation." He seems to have felt an
adjustment of the muscles ordinarily used in visual attention. At
no time was he conscious of the muscular contractions of the mus-
cles in the eyelids.
Time, .57 sec. Every fifth was accented. The series was grouped
by five and the accented click came anywhere in the group. It was
200 BOLTON :
more generally and naturally near the first place. Time, .268 sec.
The 3 -group could be suggested, and was more naturally accented
upon the first, sometimes upon the third. The 6-group was strongly
accented upon the third and slightly upon the first and fifth. With
a uniform series, the 5-group required a distinct effort and was then
accented upon the last. In general the long and complicated
groups were less differentiated; they ran together. The 6-group
broke up into two 3-groups and the 8 group into two 4-groups.
A 7-group was very difficult to get. It would run readily into an
8-group. Time, .134 sec. Every eighth was accented. The 8-group
was pleasant at this rate. When the time was .116 sec. and every
eighth accented, there was a tendency to group the 8-groups by two.
During a subsequent experiment when the time was .116 sec, the
series seemed to rise and fall in intensity with no definite grouping.
Whenever an accent was put in, it made the series irregular and
unpleasant. The series became pleasant in proportion as it was
uniform, and with this rate the subject perceived only a rhythmic
rise and fall in intensity.
Subject 16. Considerable musical talent and training.
Time, .3 sec. His first suggestion was that every other one was
stronger in intensity, the stronger one coming first in the group of
two. For a time, the subject did not discover that the sounds were
uniform. He could suggest a 4- group, in which the first and third
were accented, the first stronger than the third. It was difficult to
get a 5-group, but when the subject did, the accents were upon the
first and third. The 2-groups might be grouped by fives, in which
case the first and third 2-groups were more intense than the others;
4-groups of twos were accented upon the first and third 2-groups;
2 and 3-groups of twos were accented upon the first. Higher
groups of 3-groups as far as five were possible. The accents were
the same as for higher groups of twos. Three-groups of three were
the most pleasing. Higher groups of four were more difficult. The
accents could not be kept clear. From early childhood, the subject
has observed and taken pleasure in the rhythms in the sounds of
the fanning mill, feed cutter and other machinery. The 4-rhythm
was the prevailing rhythm with him. The puffs of the locomotive
are grouped by fours, the first and the third being accented, the
first stronger than the third. He associates the pendulum with the
2-group. With the 4-group, he associates the locomotive or a
wheel turning around, making four sounds to each revolution. The
3-group generally requires attention to keep it and a suggestion to
begin. The 5-group breaks up into a 2-group and a 3-group. The
6-group generally divides into two 3-groups. Time, .176 sec. This
rate seemed most favorable for a 6-group. It was composed of two
3-groups, the subject visualizing the pendulum which grouped the
3-groups by two. In general, the subject preferred short groups to
long ones. The shorter groups were simpler. He preferred also his
own accents to real accents. When he listened to the sound of the
chronograph, which was distinctly rhythmical, he grouped the
sounds accordingly. When he was dull and tired, faster rates were
generally more satisfactory.
The following rates were given in rapid succession: Time, .323
sec. The clicks were grouped most easily by the pendulum- swing
movement. The subject could visualize a revolving wheel which
made four strokes during each revolution and thus group by four.
Time, .263 sec. The 4-grouping was decidedly pleasant and com-
pelling. It required an effort of attention to group by three. He
visualized the locomotive wheel with the 4-group. He could group
the clicks by two with a pendulum- swing movement, but "it was
RHYTHM. 201
too fast to be real natural." Time, .137 sec. This was "a train at
full speed." The rate was more pleasant and enlivening than any
previous rate. It required very little effort of attention to get
either a 3 or 4- group. A 6 -group was easily suggested, but the
group divided easily into two 3-groups. Time, .208 sec. The 6-group
was less easy than it was with the previous rate. Time, .137 sec. The
clicks grouped readily by three or four. Higher grouping of 3-groups
by two or three required a suggestion to start, and it seemed
to continue of itself; 4- groups might be grouped by the pendulum-
swing movement. Every sixth was accented. The 6-grouping was
necessary and pleasant. The accented sound took away the effort
that had been required before for a 6-grouping. The 6-group might
be divided into three 2-groups or two 3-groups. The accented sound
always came at the beginning of the 6-group. Time, .167 sec.
Every sixth was accented. The 6-group divided easily into three
2-groups or two 3-groups. Time, .323 sec. Every sixth was
accented. The grouping was by two. The accented sound
grouped the 2-groups by three. The span for a 6-group was
disagreeable. It was too long. The accented sound might be over-
looked and the series grouped by four. Time, .167 sec. Every
sixth was accented. It was less easy to overlook the accented
click than before. The accent forced a grouping by three.
Time, .263 sec. It was most natural to group by two with the
pendulum -swing. Time, .208 sec. A 4-group was most easy.
When the subject heard the chronograph, which gave a 6-rhythm
compounded of two 3-rhythms, he grouped the sounds accordingly.
Time, 134 sec. A 3 or a 4-group was equally pleasant and easy.
The sound of the chronograph, which now gave an 8-rhythm com-
pounded of two 4-rhythms, compelled a grouping of the sounds
accordingly. The following rates were given in rapid succession
during a single experiment: Time, .268 sec. A 2 or a 4-group
was easy. A 3-group could be suggested. Time, .208 sec. A
3 -group was suggested, but a 2 or a 4-group was easier. Time, .17
sec. A 3 or a 4-group was equally pleasant and easy. There was
no preference. Time, .116 sec. The series could be grouped by
three or four. When every eighth was accented, the grouping was
by eight. At first, the 8-group divided into two 4-groups. This
disappeared, and the 8-group became pleasant and agreeable. Time,
.134 sec. Every eighth was accented. The 8-group divided easily
into two 4-groups. The span was too long. There was no satisfac-
tion in the 8-group, for the accent did not come soon enough.
Time, .208 sec. Every eighth was accented. This was distasteful.
The feeling of suspense present before was greater still. Time, .268
sec. Every eighth was accented. The suspense was still greater,
and the 8-group broke up into two 4-groups. Time, .116 sec. There
was no accent. This rate, which had given before an agreeable
8-group, when every eighth was accented, yielded to an 8- grouping.
There was a slight tendency for the 8-group to divide into two
4-groups, the first of which was more emphatic.
Subject 17. Some musical talent and training. Accustomed to
introspective study.
Time, .3 sec. In the first place, the grouping was by two, and
almost immediately and without effort it changed to a 4-group.
When each click was attended to separately, they all appeared to
be of the same intensity. Suddenly the subject began to group by
four. He felt a tendency to count it off to himself. Sometimes the
4-group appeared as two 2-groups. Then he thought there was an
irregular interval— a difference in the time of the clicks. He then
imagined that a fainter sound was heard between the actual clicks.
202 BOLTON :
Each click was grouped with the fainter sound following it, and
these groups grouped by two. Breathing seemed to accommodate
itself to the 4-group; inhalation lasted during one group of four
and exhalation during another. When every third was accented
and time .208 sec, the subject felt a strong tendency to inhale dur-
ing one group and exhale during another.
Each group is attended with the feeling of having completed a
member of the rhythm. The groups stand out as unities — as wholes
— and as each group becomes complete, there is a striving for the
next. The subject has a tendency to count the clicks by fours or
other numbers. When he attempts to suggest a 3-group, the third
click seems to repeat itself thus: 1, 2, 3, 3, — 1, 2, 3, 3, — 1, 2, 3, 3.
He succeeded, however, in getting a real 3-group by counting and
nodding the head with the accented click. When he attempted to
group by five, the accents seemed to crowd along until it brought
.six into the group. The first three clicks seemed to come in the
time of two and the rest were irregular. When he succeeded in get-
ting a 5-group, it was accented upon the second.
Time, .965 sec. The 2-group was the most natural, but it was
imperfect. Time. 1.615 sec. The subject was able by strong effort
to group by two, out the sounds seemed more naturally to appear
uniform.
Subject 18. No musical talent and no interest in music.
Time, .352 sec. This was a very pleasant rate. Other rates
seemed either too slow or too fast. By no suggestion could any
kind of grouping of the sounds be effected. The subject declared
that they were all uniform in intensity.
Subject 19. Some musical talent and in training at the public
school.
Time, .268 sec. The subject likened the series to dropping water.
It was suggested to him that perhaps some sounds were louder than
others, when he said that every fourth seemed louder. Again it was
suggested that possibly every third was louder, but the subject
would not agree to it. When every third was strongly accented,
the grouping was by three. When the accent was dropped out, the
subject returned to a 4-group. When he listened to the sound of
the chronograph, which was making a double 3-rhythm, he
grouped the sounds accordingly.
Subject 20. Some musical talent and good training.
Time, .268 sec. The sound was likened to dropping water. It was
suggested that the clicks grouped together in some way, and the
subject replied that they were grouped by four. Again it was sug-
gested that some other grouping was possible. This, the subject
said, was by three. After reflecting and counting for a moment,
the grouping was thus: 1, 2, 3, 1—1, 2, 3, 1—1, 2, 3, 1. The first ana
third were accented in the 4-group.
Subject 21. Physicist.
Time, .30 sec. The sound suggested the pendulum. A loud click
corresponded to one swing and a soft to the other. He visualized a
conical pendulum, which struck at several points in its swing and
thus grouped the sounds by other numbers than two. He seemed to
attend now to the series of clicks and then to relax and attend
again. During the " strains of attention," he might grasp three or
four clicks. A feeling of relief followed each strain of the attention.
All the muscles of the body seemed to point toward the source of
the sound. They alternately contract and relax with the successive
strains of the attention. The first click in each group was accented.
Subject 22. Some musical talent.
EHYTHM. 203
Time. .3 sec. The clicks were grouped by four. Time, .78 sec.
This rate was too slow for any grouping. It did not even suggest
the clock tick. Time, .156 sec. This rate was too fast for easy
grouping in any way.
Subject 23.
Time, .268 sec. The prevailing group was four. It was difficult
to suggest any other. The sound of the chronograph, which gave a
6-rhythm compounded of two 3-rhythms, was scarcely sufficient to
break down the tendency to group by four. The subject had
worked in the same room with the chronograph, and had become
more accustomed to the 8-rhythm than to any other which the
chronograph made.
Subject 24.
Time, .268 sec. The clicks grouped immediately by two. There
seemed to be a difference in quality. When every fourth was
accented, they were grouped by four. A longer interval preceded
the accented click. When every eighth was accented, the clicks
were grouped by eight and a longer interval preceded the louder
sound.
Subject 25. Some musical talent and training.
Subject has noticed his tendency to group objects and sounds
before the experiment. Objects passing rapidly before the eyes
are grouped by eight, those passing slower, by four, and those
passing very slowly, by two. Time, .78 sec. Every other sound
appeared to be of sharper tone than the rest. The sharper toned
click grouped with a weaker and came first. Time, .115 sec. He
grouped by eight. When he gave attention to the pulse, he seemed
not to hear the clicks coming near or just after the heart-beat.
The clicks between the heart- beats were more distinct. No
grouping of the sounds would persist long. The accented sound
in the group generally came first, but it might come anywhere in
the group.
Subject 26. Some musical talent and training. Laboratory boy.
Time, .323 sec. The most natural form of grouping was by two.
The first was accented. When he suggested a 3- group, the rate
seemed to be slower, and then the clicks seemed to be of the same
intensity. When every third was accented, the accented click
came first in the group, and was preceded by a longer interval.
In whatever position an accented click stood, it was preceded by a
longer interval. With uniform sounds the 4- group was accented
upon the first and third ; the first was stronger than the third.
Subject 27. Some musical talent and training.
Time, .3 sec. The most natural form of grouping was by two.
The first was accented. He was able to suggest groupings by three
or four. The first sound in either group was accented. By tapping
with five fingers, and striking much harder with the fifth he was
able to suggest a 5 -group. It seemed to be a matter of the imagi-
nation largely whether there was a rhythm. When he thought of
a clock or some other rhythmical machine, the series tended to
group according to the suggestion. The sound was most naturally
associated with dropping water.
Subject 28. No musical talent.
Time, .536 sec. It was possible to group the series by three, four
or five. The 4-group was most natural. From early childhood the
subject has observed the 4-rhythm in the puffing of the locomotive
especially, and in later years the same rhythm has been observed
in clocks, metronome, hammering, walking, and in all auditory
impressions that approach a regularity in sequence. The rhythm
is clearest in the sound of the locomotive. The first and third
204 BOLTON :
sounds in the group are accented; the first is generally more
strongly accented than the third. When the sounds of the loco-
motive become very rapid there is no definite grouping, simply a
periodic rise and fall in intensity. Time, .268 sec. This was
especially favorable for the 4-group, and the 3-group could not be
easily suggested. Time, .208 sec. The 4-group was most natural.
It was possible to suggest a 2 -group by striking heavily on every
other sound. The grouping, however, was very monotonous. Both
the 3 and the 5-group were very difficult. Time, .268 sec. Although
the 3-group was difficult at this rate before, it could be easily sug-
gested this time. Time, .17 sec. The clicks were grouped by four
and the 4- groups tended to group by two with the pendulum- swing
movement. If the grouping was held down to a plain four, it be-
came unpleasantly monotonous. Time, .134 sec. The series tended
to appear in the form of a periodic rise and fall in intensity. The
periods were about equal to the time of an 8-group, and with a slight
voluntary effort the series grouped by eight. The 8-groups tended
to group by two with the pendulum- swing movement. During a
subsequent experiment with the same rate, the subject felt a ten-
sion in the eye muscles which grouped the series by eight ; four
sounds occurred during the upward movement and four during the
downward.
Subject 29. Some musical talent and training.
Time, .268 sec. When the subject thinks of a clock the series
groups by two, but when he thinks of hammering, the clicks
appear to be of the same intensity. He could suggest a 3 or
4-group, but the 2-group was most natural. Time, .208 sec. He
finds it easy to count almost any rhythm as far as nine. The longer
rhythms tend to divide into shorter ones. The subject found it
difficult to keep from thinking of a clock tick, which suggested the
2-group. Time, .17 sec. The subject still grouped by two and
thought the rate seemed to be faster when he grouped by two than
when he suggested other groups.
Subject 30. No musical talent.
Time, .268 sec. By no suggestion was it possible for the subject
to effect any kind of grouping of the sounds. It appeared as a dead
monotonous series, with which he could not avoid the association
of a pile-driver.
Many other persons who simply came in as visitors, were experi-
mented upon with results which confirmed the foregoing records.
No especial account was taken of them. More than fifty persons
in all were experimented upon, and only two failed to effect some
kind of grouping in the clicks which they heard. In general it may
be said that the younger and less educated yielded more easily and
quickly to the suggestion of a rhythmical grouping.
The first point in the preceding records to which attention
is called is the rhythmical grouping of the sounds. The
grouping was the same in every case. It was accomplished
by accenting regularly certain sounds more than others.
The weaker or less accented sounds seem to run together
with the stronger, and to form organic groups which are
separated from one another by intervals which are apparently
longer than the interval which separates the individual clicks.
Such rhythmical grouping has been observed frequently at
other times by many persons. Several of the subjects testify
RHYTHM. 205
to have known of their tendency to group the puffs of the
locomotive, even in early childhood, and they have taken
great delight in it. With us this habit of grouping the puffs
of the locomotive when it was starting slowly or pulling up a
grade became so strong, even in early childhood, that it led
to all kinds of speculation as to the cause. The puffs are
grouped by four. The first and third are accented, the
first generally stronger than the third. No other grouping
ever seemed possible until it was found in the experimental
work that the tendency to group by four was only a habit or
association. The puffs of a locomotive may now be grouped
by two or three, but the association of the drive-wheel making
one revolution to four sounds renders any other form of
grouping than by four difficult. When the engine runs very
fast, the sounds seem to rise and fall in intensity at regular
intervals.
A kind of rhythm is also observed in the noise of mill-
wheels. The winnowing machine and feed cutter, such as
are found upon many farms, produce a rhythmical sound
which few persons fail to observe. Long association in early
childhood with such rhythms stamps them upon the mind so
firmly that they become a mental habit. Children either
fancy or perceive rhythms in many sounds ; they indicate
this by their attempts to reproduce the sound of machinery
or of locomotives. Some railroad engineers believe their
engines sing tunes. The same engine under like circum-
stances always sings the same tune.
Several experimenters have also observed this same group-
ing of rhythmic sounds. In the work undertaken by
Dietze 1 in Wundt' s laboratory upon the Umfang of con-
sciousness, this rhythmical grouping of the sounds of the
metronome was observed and employed to determine the
length of the mental span. The grouping was accomplished
by intensifying voluntarily certain sounds and subordinating
others to it. By grouping the sounds first by eight and then
the groups of eight by five, it was possible to grasp forty
sounds. Wundt says it is impossible to restrain this group-
ing absolutely. It may be confined to a 2-group, beyond
which it cannot go within certain limits. Four sec. is the lower
limit, and .11 sec. is the upper limit. The most favorable
rate is .2 to .3 sec. Wundt refers this grouping to the ripen-
ing of the concept on the wave of apperception. As we shall
see later, it is possible to restrain this tendency to group
sounds. The difficulty was with Wundt' s apparatus. The
two sounds heard during a complete swing of the pendulum
a Wundt, Physiologische Psychologie, Vol. II. p. 73.
206 BOLTON :
of the metronome are not of the same intensity or quality,,
and hence the impossibility of restraining the grouping by two.
Angell and Pierce, * in their experiments upon attention,
state that one subject noticed a rhythm in the sounds with
which he felt a tendency to muscular contraction — nodding of
the head and beating time with the fingers.
In neither of these experiments could the experimenters be
sure that there was not some difference in the sound which
would suggest a rhythm. The importance of an absolutely
uniform series of sounds cannot be too strongly insisted upon.
A difference in sounds which would ordinarily remain
unnoticed, is sufficient to suggest a rhythm. This will be
seen when we come to discuss the voluntary changes of the
grouping and the ease of suggesting such a change. In the
present experiments the greatest precaution was used against
any variation in the sounds that would suggest or impose a
grouping. The only possible source of such a variation
would come from a difference in the resistance between the
mercury and platinum. If the mercury were dirty or the
platinum points were not sufficiently immersed to form a
good contact, or the mercury were to adhere to the points as
they were withdrawn, a difference in the intensity of the
sound might be heard. The mercury was carefully cleaned
every few days, or fresh mercury put in. The platinum
points were filed smooth and kept brushed. Strong elastics were
attached to each key, so that when the keys were released there
was no delay about reacting. If then there were any varia-
tions, since there were five sets of keys, it ought to recur
every fifth sound ; but as a 5-rhythm was always found very
difficult, and a 2, 3 and 4-rhythm easy, we have strong
ground for believing that any variations except those which
were intended were so small as to have no influence upon the
rhythmical grouping. We have, then, the testimony of all the
subjects that the clicks seemed uniform in intensity.
Subject 2 always heard a uniform series for a time after a
change of rate, or at the beginning of a new experiment.
His tendency to group was so strong that he could avoid it
only by imagining some one pounding in the distance, or
some objective thing that was perfectly uniform. Sub-
ject 3 did not feel any tendency to group the sounds until
after he had tried several suggestions. Subject 9, taking a
critical attitude, was inclined to believe for a time that the
clicks were all of the same intensity. After a few moments
it required an effort, which was like " looking long into the
American Journal of Psychology, Vol. IV. pp. 534 and 539.
RHYTHM. 207
future, " to avoid a grouping. " I find no rhythm, " he says,
" as long as I hold my breath and stick to it. 7? When sub-
ject 11 gave close and critical attention to the sound, there
was no grouping. In order to get a notion of a rhythmic
series — one of uniform intensity — subject 12 turned his atten-
tion " backward " and saw a series of images to which he
was adding one all the time. He throws his attention upon
what comes, and studies the nature of the sounds to see if the
timbre remains the same. Subject 1 7 says that when each
click was attended to separately, they all appeared to be of
the same intensity. He said he experienced no such difficulty
in avoiding a rhythm as the statement of Wundt had led him
to suppose. Subject 25 could group the sounds, but he was
more inclined not to do so. If he suggested a grouping, it
did not persist. Subject 27 found it more natural to associate
the sound with dripping water. Subject 29 made the series
appear uniform when he thought of hammering. Subjects 18
and 30 could not effect any grouping at all. Upon this
evidence we may safely rely upon having secured a series of
impressions that was uniform for sensation. It is also true
that though the rhythmical grouping of a series of uniform
sounds is difficult to avoid, this tendency may be restrained
within the limits spoken of by Wundt. Our own experience
tallies with those above. When the attention is directed to
each single impression, and an attempt made to study the
timbre, it is possible to restrain the rhythmical grouping of
the sounds. But when the series is attended to as a whole,
this grouping takes place involuntarily.
The character of the sound employed in the experiments of
Dietze differed greatly from that used in these experiments.
The click of the telephone is about as simple and instantaneous
a sound as it is possible to produce. The plate in the
telephone vibrates a very short time. For that reason its
chief characteristic is intensity ; it does not persist long
enough to establish its pitch and timbre. The mind has very
little to work upon. It can construct variations only in
intensity, for which reason the carrying power is greatly
reduced. The sounds can be subordinated with respect to
intensity only, and unless great intensive variations can be
made, the mind will lose its grasp, and the grouping break
up into single impressions. This phenomenon was observed
several times, and in particular by subject 15. The sound of
the metronome which Deitze employed is full and rich and
has greater carrying power. Any experiments upon the
carrying power of the mind must take into consideration the
character of the sound. Dietze was able, by strong voluntary
effort, to carry the grouping much farther than any subject in
208 BOLTON :
this experiment was able to do with the clicks of the tele-
phone. The explanation is to be found partly in a differ-
ence between the two sounds and partly in a different method.
The subject in these experiments was requested to group the
sounds, not by voluntary effort, but only so far as it was
found easy and spontaneous. There was no attempt to force
the grouping as far as possible, or even to force the grouping
at all. It was the spontaneous and involuntary grouping
that was studied.
In a study like this, which is purely introspective, the
experimenter must rely upon the integrity of his subjects.
There is and can be no test of the accuracy or truth of the
results, except the uniformity which they show. If, however,
each subject is unaware of the object to be obtained by the
experiment, and of the opinions of every other subject, and
renders his judgment without any interest in the results or
without any preconceived notions of the experiment, the
judgments are no more subject to error, and have about
the same value as judgments in psycho-physical experi-
ments. Certain attitudes, habits, and characteristics of mind
do, however, affect results in certain ways which are injurious
to the experiment. Some attention was paid to the attitude
and method of the subjects in making judgments. A few words
in regard to this may not be out of place. There are three
classes of psychological subjects. The first includes those
persons who yield immediately to any suggestion that is
offered. This attitude results, then, from a social practice.
In society, people do not wish to antagonize others. They
instinctively give assent to any opinion. In an experimental
investigation, if the operator will just give the slightest hint
of his theory or preference they will add the weight of their
opinions. If the operator leads them into giving an opinion
which is opposed to his theory, " consistency becomes a
jewel ;" they stick to their opinion stoutly. If the experi-
ment shows plainly that they are wrong and it is preposterous
to hold such a view, they make a compromise with their
former position and try to excuse themselves for having been
led astray. They remain respectfully silent afterward and
avoid, if possible, giving an opinion. If they are forced to
make a judgment, they do it tentatively; they are not sure.
Of a number of possible views they cannot make up their
minds which is the correct one. They generally hair-split
until they find out someone's opinion and then agree with that.
The second class of subjects includes those who take a mod-
erately critical attitude. They are concerned in others'
opinions in so far only as other opinions suggest different
points of view. They give their own opinions when they
RHYTHM. 209
have considered all the phases of the experiment that are sug-
gested to them. They are unconcerned about the outcome of
the experiment. They are not dogmatic ; they might have a
different opinion under different circumstances or with further
consideration. In the light of the evidence before them, they
hold to a certain view.
The third class includes those persons who are excessively
critical. They incline always to an opposite view. The
experiment is not conducted properly to suit them ; they are
not in their best mood for judgment. They are sure to take
ground against some one's opinion. If they cannot get any
clue to others' opinions, they are doggedly silent or quibble,
and refuse to answer except they qualify their answers to
such an extent that the answer means nothing. This class of
subjects is intellectually dishonest. If they are compelled
to answer, they indulge in hair-splitting differences between
their opinions and those of some others.
When the experimenter is compelled to rely entirely upon
the judgment of his subjects, he must study them carefully
and use the opinions of certain subjects in so far only as he
finds that they harmonize with the general results. It is a
fact which every psychologist must understand that certain
classes of persons are incapable of introspection. The first
class to which we referred are unfitted, because of habits of
too free judgment and of always agreeing with others. The
third class are rendered unfit for introspection from habits of
too free judgment in regard to matters that concern themselves,
and from an unnatural bias toward the negative. They are
inclined to make too much of their individual opinions. In
making out the results, the investigator cannot rely much upon
individual opinion. Where there is almost perfect uniformity,
the results may be given in tabulated form ; but a large space
must be given to merely individual opinion.
We have then to inquire first in regard to the certainty of a
rhythmical grouping of a series of absolutely uniform sounds.
The point does not need argument ; the preceding records show
how strong is this tendency. Only three out of fifty or more
persons tested would agree that it was easier to hear each
click separately. In addition to the records given above,
several subjects were asked to give a written statement of
their impressions of the experiment. In one case definite
questions were asked in writing.
(A) "As far as I can recall my impressions at the different
occasions on which I listened to the series of sounds from
your apparatus, they appeared to me always as a sequence of
groups containing the same number of elements. The excep-
tional cases where the impression was that of a sequence of
210 BOLTON :
single sounds, were those in which the period of the sequence
was at its longest. For any given rate there was in general
one certain number of elements of which the groups more
naturally consisted than any other: but I found, too, that the
sequence took on instantly the character of almost any other
grouping that was suggested, whether by word or sound. As
to the psychological nature of this phenomenon of grouping,
it is a difficult matter to give an opinion. I found the effort to
determine whether or no there were any recurrent differences of
sensation in the sequence a great strain upon the powers of
attention. The grouping had in general the appearance of
being forced on the mind by the sounds rather than that of
being imposed on them by it."
(B) "A series of clicks may be given in such manner that
by giving the closest possible attention they seem to be
uniform both as to intensity and interval. This degree of
tension (of attention) can, however, be maintained for only a
few seconds. When the attention is moderate, the clicks tend
to fall into rhythmic groups, the number of clicks falling into a
group varying with the rate of the clicks. Slower than a cer-
tain rate no rhythm is felt. With more rapid rates two clicks
form a group, the accent falling on the first and an interval
occurring after the second. Faster still, four clicks form a
group with accent, primary on the first and secondary on the
third, and an interval after the fourth. This seems a very
pleasing rhythm to me, more so than any other. A still more
rapid rate gives eight in a group. This becomes visualized
quite strongly in my case. It is exceedingly difficult for me
to hold the series of clicks out of some of these rhythms.
They fall into one or the other types (according to rate)
almost irresistibly. At some rates I was able to get a
3-rhythm, accented strongly on the first."
(C) " With regular ticks within certain limits, I do not
perceive them as distinct separate ticks, but from the first I
group. With slower rates, the grouping is two by two, which
passes very easily into four, subdivided into two. With
faster rates, the tendency is 4o perceive the grouping into
fours, divided into two, or to perceive the grouping into
threes. The quicker the rate, the larger the number of ticks
entering into the groups up to about six. Below the lower
limit, the ticks are first perceived separately with a tendency
to fall into twos, this tendency decreasing as the rate
decreases. Above the upper limit, the grouping becomes
vague and the tendency is to perceive the ticks as separate
and individual. In general the grouping can be changed
within certain limits."
RHYTHM. 211
"The groupings influence one another. There is a tendency
to become habituated to a grouping. A grouping heard in
one rate is likely to repeat itself in a subsequent rate. It is
difficult to be perfectly passive when one knows he is to find
a rhythm.' }
(D) "It seems to me easier to group the clicks unless they
are very slow ; but I do not find it so difficult to perceive them
singly as I should have inferred from Wundt's remarks on
Dietze's experiments. Having now tried many times when
the grouping was strongly present, subjectively (voluntarily)
or objectively, I think I am a little more inclined to discover
groupings. It seems to me that I do not lengthen, but rather
intensify one or more of the sounds. Perhaps, however, the
change is more in quality than in intensity, or perhaps an
accompanying impulse of the diaphragm, stress in the men-
tal counting, etc., etc. Possibly, however, I do also
lengthen the stressed sound at the same time; but the
lengthening is not so clear as the stress. I infer from my
experience as a subject that the rates from 1 or 2 per sec, up
to 6 or 8 per sec, are best; probably about 4 per sec. being
the best. The fast rates are better for groupings by four, the
slower for groupings by two. Three-groups, 5-groups and
higher groups do not occur spontaneously with me, though
3-groups are not hard to start by counting. Perhaps 2-groups
go easiest of all with me. There is a sense of expectation of
' hope deferred ' when the rate is too slow — or, at least, a
feeling of ' too slow,' like traveling in a slow train, although
you have plenty of time.
" This probably increases with the length of the group. The
span of the respiratory rhythm is exceeded, and instead of
being able to tell off a whole foot of the rhythm with one
breath, several breaths intervene between those that mark the
accented sounds. With small groups and rapid rates there is
a feeling of hurry. The motion is too quick and short. There
is none of the repose — the swaying, the grace, the easy fulfill-
ment of expectation that a slower rhythm possesses."
This rhythmical grouping was a series of efforts to attend
to the sound. The grouping results from a sequence of acts
of attention. When the attention is directed to the sensation,
it lays hold upon the first impression with great force and
makes it the sole object of consciousness. If this were the
only sound, the attention would turn to something else, but
as succeeding impressions follow before the first wave of
attention has subsided, they are seized upon with less force
than the first impression, and are subordinated to it in differ-
ent degrees according to the strength of the apperceptive act.
Subsequent waves of attention follow the same process as long
212 BOLTON :
as the will directs the attention to the phenomenon. The
attention accommodates itself to a certain number of impres-
sions, which fall easily within the period of a wave, providing
there is no objective difference in the impressions. If there
is a regularly recurrent difference, this becomes the signal for
a new act of attention, providing only that the span does not
exceed or fall much under the normal period of a wave. If
this recurrent difference follows at too great intervals, the
attention breaks up the span in two portions, the one more
emphatic than the other. If it follows at too small intervals,
these periods fall together into group, first of two and then of
larger numbers. The too great interval is marked by a feel-
ing of suspense, and the too short interval by a straining
after something more.
The number of uniform elements which may enter into a
member of the sequence is not determined wholly by the time
interval which separates them. Previous mental habits and
associations influence the number of elements in the members
of the sequence. All individuals are more habituated to two
and its multiples than they are to three. There are also many
associations which will suggest groupings by two and four;
All ordinary muscular movements follow a rhythm of two.
The associations of four are far more frequent than those of
three. For this reason to a large extent, groups of two and
lour prevail. Several subjects have described this effort of
attention in a manner which deserves notice and which shows
very well the nature of the act.
Subject 7, speaking of his grouping by eight, says he is not
able to " round up" until he comes to eight. There was
feeling of completeness about the 8-group with a certain rate*.
Subject 9 says there is a slight feeling of muscle tension in the
ear, sometimes in the back of the scalp. He attends, relaxes^
and attends again. There is an innervation of the muscles
connected with attention. Subject 12 describes his feeling
about the grouping as a tendency " to go back " when he has
heard three or four clicks, as the case may be. This is a
" mouthful " — a unity, and, when he has one, he seeks to get
another. Subject 11 describes his feeling as a series of efforts
of attention. He grasps and grasps again Subject 17 says
each group is attended with a feeling of having completed a
member of the rhythm. The groups stand out as unities — as
wholes — and as each group becomes complete, there is a
striving for the next. Subject 21: "I attend now to the
series of clicks, then relax and attend again. During the
strains of attention, I may grasp three or four clicks. A
feeling of relief follows each strain of attention. All the mus-
RHYTHM. 213
oles of the body seem to point toward the source of the sound.
They alternately contract and relax, "
This is the rhythm in the attention to which a reference was
made above. The view taken, then, was that only one undi-
vided state of consciousness might arise during each pulse or
wave of attention, and that the number of objects which can be
grasped in that state must form an organic unity or be pre-
sented as a single object — have the appearance of a unit.
A given number of auditory impressions within certain
time limits, when presented in such a way that there is a
kind of subordination among them with respect either to
time, intensity, pitch or quality, or with respect to any
two or more of these properties, always stand as a unit in
consciousness. They form an organic unity which is the
essential condition of a number of impressions entering into
a state of consciousness. If such organic unity does not exist
and it is possible to make it, the mind imposes such an
arrangement upon a given number of the elements that they
may enter into a state of consciousness. The essential con-
ditions of forming such a unity among sounds is a regular
temporal sequence within limits which shall be named here-
after, and perfect uniformity in intensity, pitch and quality.
Begular variations within limits with respect to intensity,
pitch or quality, or to any two, or to all of these together,
will effect a subordination among them sufficient to constitute
an organic unity. There is a temporal limit within which
these variations must occur in order to form such a unity.
The test of how many auditory impressions might be
grouped together was the ease and pleasure which the subject
found in doing so. If he were compelled to keep up a con-
stant suggestion of a particular number in order to group the
clicks so, noaccount was taken of it. If, after suggesting a
grouping, it should persist until some other suggestion was
made, the rate was considered favorable for that form of group-
ing. The subjects have described some groupings as most
natural, easy or pleasurable, and others difficult or displeas-
ing. The groupings which were spoken of as natural, easy
or pleasurable, are gathered together in the following table,
with the time, to determine what rates have been found best
adapted to the different forms of grouping.
In the following table are brought together the judgments
of all those subjects with whom extensive observations were
made. The number of the subject is given in the first column
at the left hand, and in the columns to the right are given the
rates in thousandths of a second, at which a certain form of
grouping was found pleasant and easy. The designations at
file top of columns 1, 2, 5, 8, 10 and 13 are sufficiently clear.
214 BOLTON :
The others require further explanation. In column 3 are given
those rates at which the subjects found a 2-group more easy,
but there was a straining for a larger group, or the 2-groups
seemed to group by two. The rate was a little fast for a
2-group, and yet it was not more pleasant to group by four.
In the same way certain rates were found at which a 3, 4, 6
or 8-group was easier than any other, but it was a little too
fast for simply grouping by these numbers, and hence the
groups tended to group by two. This was generally spoken
of as the " pendulum -swing movement. " Still other rates
were found at which a 4, 6 or 8-group was more pleasant,
and yet the rate was too slow, and the group tended to divide
into two^smaller groups. In column 15 are given those rates
at which there was no distinct grouping — simply a periodic
intensive change in the series. Bates at which there was no
appearance of a group are given in column 16.
Multiplying the average rate for each form of grouping by
the number of clicks in a group, we get as the length of groups :
Lower limit for no group, 1.581 sec. Average variation, .29 sec
Average length of 2-groups, 1.590 " " " .328 "
" " " 3 " 1.380 " " " .204
" " " 4 " 1.228 " " " .068
" " " 6 " 1.014 " " " .028
" " " 8 " 1.160 " " " .025
The foregoing table shows that the lower limit for the
rhythmical grouping of sounds is near 1.58 sec. Some sub-
jects are able at times to group sounds that are separated by
this interval, but as a general rule spontaneous grouping has
ceased. The records give several instances where the sub-
ject has visualized the pendulum with this rate, but he had a
feeling that the pendulum reached its full swing before he
heard the click. The upper limit at which spontaneous
rhythmical grouping ceases cannot be far from .115 sec.
Several subjects declared their inability to make definite
groups at a rate less than this. Others perceived only a
periodic rise and fall in the intensity of the sound ; there was
no definite grouping.
Between these limits there was some form of rhythmical
grouping which depended in a large measure upon the rate.
The average of all the rates at which a grouping by two was
found easy and natural has been taken and multiplied by two
to find the average length in time for the 2-groups. The
same has been done for groups of three, four, six and eight.
The averages for groups of all forms are found not to differ
greatly, when we consider certain facts which influence the
length of the group. The average length of 2 and 3-
groups is somewhat greater than the average for groups of
1
2
3
I
4
P
5
6
6
7
Is* 3
8
9
6
10
11
d
12
13
14
d
15
16
22
i
a
5
4^
a
S3
t
1
1
3
&
s£
0Q.fi 05
m
0.
S£
S.*A
s.
a&
Pi
IS
a-sa
&
i&
°*8
i
go*
2s
3® 5
2-2 2
g
u 3
3 ® 3
§
§§<
$
3ft
2s
s « 5
2s2
1
So,
2s
sis
&
■§
o
4Qci»
3
4o
ooQ4
9
00
ooO
CO
9-
2>&
2
9^
o
I
1152
1140
.570
.230
.323
.208
.167
1000
.323
2
1.002
.323
.323
.263
.263
.323
.263
...
....
.208
.167
3
2 304
.323
.500
.268
.760
.760
.500
...
....
.760
.167
1.140
1.140
.230
4
1 660
1.072
1072
.646
.646
.536
.536
.323
.288
.268
.134
.208
.268
.323
.170
.116
.156
.134
.116
5
.780
1.440
.288
,420
.323
.353
.208
.156
.183
.208
6
1440
1.660
.300
.780
.353
.780
.300
.353
.288
.167
.156
.288
.134
.134
.156
.156
.137
.156
7
1 140
.780
.300
156
8
1 440
.300
167
.137
156
.208
9
2
1.440
.780
.760
.540
.268
.300
.134
.134
.268
.156
.166
.116
.116
10
1.440
.450
.268
134
.116
134
268
134
.156
.268
.300
.116
.353
.780
.300
11
1.67
.268
.500
.268
.167
.137
.115
...
.134
.208
.115
.250
.263
.156
.268
.115
.323
12
1.660
1.56
.156
.156
.116
«...
.156
.156
.115
.115
... •
1.440
.760
.300
.115
13
.285
.288
.285
340
115
... •
....
.352
.352
.352
.780
.288
.352
.969
.268
.208
14
.780
1.440
780
300
.156
.285
.285
15
.300
.268
...
.134
.116
.116
....
.268
.116
.134
16
1.660
1.720
288
.156
.115
.323
.263
.167
17
1.615
.789
.965
.969
.849
.268
.352
.624
.789
.352
.307
.307
.500
.167
.208
.167
.624
.501
.167
.789
.192
.208
19
.268
.268
.300
.268
;;:::;
•••
.::;
•••
:;;;:;
....
;::;:;
::::::
20
22
.780
.156
28
1.938
1.672
.536
.208
.167
.134
.167
.134
.323
.156
.137
.134
....
.969
.323
.170
.116
.116
.323
.156
.115
....
.700
.323
.134
.969
.208
1.072
.208
.268
.536
Average
1.581
.795
.526
.542
.307
.183
.134
.145
.125
.460
.204
.149
.161
.169
.137
.127
Average
Variation
.29
.328
.204
.068
.043
.025
.009
.028
.028
216 BOLTON :
six and eight. The explanation for this is to be sought in
the fact that persons are more accustomed to rhythms of two
and four than to the longer rhythms.
The average variation for 2 and 3-groups is greater
than for groups of four, six and eight. The associations with
the 2-rhythm are far greater than with any other, and these
associations tend to suggest the 2-group where it would
not otherwise occur spontaneously. Long experience with
clocks that vary greatly in their rates of ticking has much to
do with the wide limits within which the 2-rhythm is possible.
The 3-group is a more rare form of grouping, and only a few
subjects succeeded in suggesting it easily. For that reason
it is not surprising that the average time of the 3-group
should differ greatly from the others. Then the power to
carry one or two impressions in the mind is greater, and they
can be held longer. The actual span for two clicks in a given
time is only a little more than half the span for eight clicks
in the same time. In the first case almost half the interval
is a pause between the groups, and in the second the pause
takes up less than one-eighth of the interval.
There are several facts, as the records have shown, that
tend to make the length of groups vary. Several subjects
were predisposed to groupings by four. This number has had
a peculiar charm for one from early boyhood. It was his num-
ber in school, and ever since, objects that were grouped by
four, or that could be grouped by four, have had an especial
attraction for him. Four impressions, of whatever sort,
always arrest his attention. For this reason he attempted to
group all rates by four, even though it required strong effort
to do so. With two exceptions, all subjects had the prevail-
ing tendency to group by four. A second fact, which in-
fluenced probably the grouping to some extent, was that when
a subject found it easy to group a given rate by four he
became somewhat habituated to a 4-grouping, and was in-
clined to group the succeeding rates by four, unless they
differed greatly. If a very slow rate followed a faster one,
which had been grouped by three or four, the subject tended
to imagine intermediate clicks between the actual clicks, and
still to group by three or four, as the case might be. ( See
the records of subjects 1, 6, 12, 13 and 17.) Taking all the
forms of grouping together, the average time is taken to indi-
cate the normal period of a wave of the attention which does
not exceed greatly one second. A spontaneous effort of the
attention, or with Wundt a wave of apperception, endures
about a second or more. We do not, however, hold that
there is an absolute psychical constant, even for the individual.
No other fact is more certain than that the condition of the
RHYTHM. 21T
subject, as regards fatigue and previous engagement, has
much to do with the rate at which a certain group is found
pleasant and agreeable. (See the records of subjects 1 and 3. )
Before leaving the subject, let us call attention to the
averages for groupings that are intermediate between two
and four, four and eight, and three and six. The averages
for the rates at which these groups were observed lie between
the average rates for the groups between which they stand.
A further method of testing the normal length of a sponta-
neous effort of attention was sought in this way. An accented
sound every sixth or eighth was introduced into the series,
and a number of different rates were tried, until one was
found at which the group seemed most pleasing and natural.
If the rate was too slow for easy grouping, the subject per-
ceived a feeling of suspense. A slower rate still, caused the
group to divide into two parts, or at least the subject felt a
tendency to divide the group. A still slower rate generally
caused the long group to disappear entirely, giving place to
a number of small groups which were equal to the long one.
If the rate were too fast for easy grouping by six or eight,
the groups tended to group by two with a kind of pendulum-
swing or wave-like movement. Before trying a subject upon
an accented series of six or eight, he was given a number of
rates with uniform clicks, beginning with a slow rate. The
purpose was to determine to how great an extent the form of
grouping changed with different rates, when they were given
in close succession. The results of the experiment with a
series of uniform rates are given in the first part of the fol-
lowing table. The results of the experiment when every sixth
click was accented are given in the second part, and when
every eighth click was accented, in the third part :
Unaccented Series, Part I.
Subject 7.
Subject 16.
Subject 3.
Subject 4.
Subject 10.
Subject 13.
Subject 16.
Subject 9.
Subject 17.
Subject 5.
.323
(.268)
4- group.
Tends slightly
to 8-group.
2 -group.
3-group.
.268
4- group.
Pleasant.
4- group.
Pleasant.
2 -group.
Groups rise
and fall.
2 or
4- group.
4- group.
Tends slightly
to double
4 -group.
.78
2 -group.
.263
4- group.
Strongly
tends to 8.
4- group.
Double
3-group.
.208
4 -group.
2 or
4- group.
.353
4 -group.
.208
8-group.
Pleasant.
.17
4- group.
Confused.
.156
6 -group.
Pleasant.
4 -group.
Groups rise
and fall by
two.
3 or
4 -group.
Wavers
between
4 and
8-group.
.167
Groups not
separated.
Confused
feeling.
6 -group
Not so easy as
with the
following rate
3-groups.
Groups by
two.
.134
4- group.
8-group.
Pleasant.
4-group.
Strongly
towards 8.
8-group.
Illusive.
.183
8-group.
.137
More
confused
feeling.
Accented
double
6-group.
6-group.
Groups by
two.
.116
4- group.
Groups by
8-group.
Groups by
two.
8-group.
Divides into
two 4-groups.
8-group.
Divides into
two 4-groups.
8-group.
.156
8-group.
Accented 6-Group, Part II.
.323
.263
.208
.167
.137
Subject 6.
4-group.
In spite of
accent.
6-group.
Suspense
tends to
3-group.
6-group.
Difficult.
6-group.
Pleasant.
6-group.
Groups by
two.
Subject 11.
6 -group;
Tends to a
4-group.
6-group.
Pleasant.
6-group.
Groups by
two.
Subject 16.
6-group.
Two 3-groups.
Span disagree-
ably long.
6-group.
Divides into
two 3-groups.
6-group.
Pleasant.
Pendulum-
swing
movement.
RHYTHM.
Accented 8-Groups, Part III.
219
Subject 4.
Subject 10.
Subject 11.
Subject 13.
Subjects
.268
4- group.
4- group.
8-group.
Really
4-groups.
Great
suspense for
8-group.
.208
4- group.
Accents
disturbing.
8 -group.
Not complete.
Two 4 groups.
8 -group.
Distasteful
feeling of
suspenss.
.17
4- group.
8-group.
Requires
effort.
8-group.
Accents
disturbing.
8-group.
Divides into
two 4-groups
.134
Two
4-groups.
8-group.
Not
animating.
8-group.
Not pleasant.
8-group.
Pleasant.
8-group.
Divides into
two 4-groups.
Span too
long.
.116
8-group.
8-group.
Harmonious.
8-group.
Pleasant.
8-group.
Groups by
two.
8-group.
Pleasant.
With the unaccented series, the 6-group was found natural
twice near the rate .167 sec. When every sixth sound was
accented, the most pleasant rate for the 6-group was .167 sec.
At the rate .137, the 6-groups group by two. At the rate
.208 sec, they were difficult to grasp. At slower rates, there
was a feeling of suspense, or the group tended to divide into
two 3-groups, or the subject was more inclined to group by
four in spite of the accent. According to this, the 6-group is
found most natural and pleasant at the rate .167 sec. By
multiplying this by six, we will get as the time limit for the
6-group 1.002 sec.
With uniform series, the 8-group was found most natural
and pleasant, once at the rate .208 sec, once at the rate .134
sec. and twice at the rate .116. When every eighth was
accented, the 8-group was found most pleasant at the rates
.134 and .116 sec The average rate for all is .130 sec,
which, when multiplied by eight, gives 1.04 sec, the time
limit for the 8-group. The difference between this and the
time for the 6-group is very small, and at the same time they
agree very well with the times for the same groups in the pre-
ceding table. The general fact of certain rates being better
adapted to certain forms of grouping is pretty well established.
This adaptation of a particular form of grouping to a certain
rate depends upon the fact that the length of the group cor-
responds to the normal period of a wave of attention. The
lack of adaptation results from cutting short the normal wave.
For a fuller account of the different states of feeling arising
220 BOLTON :
with different rates for a certain group, the reader is referred
to the records of the experiments upon subjects 10, 11, 13
and 16.
The conscious state accompanying each wave of attention
grasps together or unifies all the impressions that fall within
the temporal period of a wave. As the result of a series of
attentive efforts, a series of auditory impressions takes the
form of a sequence of groups. This rhythmical grouping is
due to the unifying activity of the mind ; it is an attempt to
conceive a series of sounds in a simpler form. When the
mind acts upon a continuous series of auditory impressions,
it groups all the impressions that fall within the period of a
wave of attention, and conceives them as a single impression
or a unity. Each succeeding wave groups a like number, so
that the series is conceived in the form of groups. If the
single impressions are separated by a greater time interval
than the normal period of a wave of attention, each impression
stands alone as the sole object of consciousness. But what
becomes of the series when the rate is too fast for rhythmical
grouping ? A partial answer is to be found in the fact that
the clicks show a regular periodic rise and fall in intensity.
There was no separation among the groups ; no definite
number of impressions constituted the group. The view to
which least objection can be offered, but which is unsupported
at the same time by any positive evidence, is that when the
sounds become too rapid to find expression in muscular con-
tractions of any kind, they can be no longer separated from
one another as simple impressions.
The most rapid rate 1 of voluntary control is about ten per
second. This periodic rise and fall in the intensity of the
clicks simply marks the waxing and waning of the attention.
The changing intensity of the sounds indicates the changing
degrees of clearness in the conscious state. If the clicks are
separated by more than one- tenth of a second, the groups are
separated by an interval ; at least, there is the feeling of an
interval. Below this limit of one- tenth of a second, the clicks
preserve their individual character. They do not fall suffi-
ciently near together to appear continuous. They preserve
their temporal succession, and as before appear in different
strengths according to the degree of clearness in the conscious
activity. The conscious state, which seems to ride upon the
crest of the wave — that is, appear when the attentive effort is at
its strongest — fades gradually and conceives the last elements
in the group with less clearness than the first, and with the
coming of a new wave of attention, the first impression is laid
1 " Some Influences which Affect the Rapidity of Voluntary Move-
ments." F. B. Dresslar. Am. Jour, of Psychol. Vol. 4, p. 516.
RHYTHM. 221
hold upon with great force, and appears stronger in contrast
with the last in the preceding group.
This rhythm in the attention, and hence in conscious activ-
ity, finds its counterpart in the activity of the nerve cell,
which we have seen reason for believing was a series of explo-
sions — an alternation of periods of activity and periods of
repose.
The subject invites speculation, but we forbear except to
offer the further supposition that with rates slower than ten
per second, the interval or pause between the rhythmical
groups marks a period of perfect quietude in the cell. When
the rate reaches ten a second or more, there may still be a
period of absolute inactivity, but no less interval than a tenth
of a second can cause a real break in the conscious state or
no less interval becomes an " object of consciousness." The
change from one state of consciousness to another is repre-
sented by the reversal of a muscular movement. If between
two impressions there is not sufficient time or time equal to
the reversal of motion in a member, there is no consciousness
of an interval between the impressions. The thought of the
interval is a deduction and not a sensation or conscious fact
— a fact revealed by the immediate conscious state itself.
The conscious state disappears when the activity in the cell
ceases ; and when the will directs the attention to the series of
impressions, the conscious state tends to disappear when it
has effected all the subordinations that are possible among the
impressions that fall easily within the normal period of a
wave of attention.
Another phenomenon, which was observed by several sub-
jects and by the experimenter at different times, was the
apparent slowing up of the rate. The feeling was one of
extreme suspense, and was described as "awful" and
"dreadful." There was no apparent regularity with which
the slowing up occurred and no definite time that the feeling
lasted. The only fact which was observed concerning it was
when the attention was diverted, the feeling disappeared.
Several suppositions occur to us as explanations, but none of
them seems to be completely satisfactory. It appears to be
more in the nature of fatigue, but it is not clear why the
impression should seem to be separated by longer intervals.
The general iact, however, of time passing more slowly, when
one is suffering from fatigue, has been observed frequently. A
more probable supposition is that it indicates a kind of
rhythm in the voluntary effort which directs the attention to
iihe source of the sound.
We have now to ask what is the inherent nature of a
rhythmical group, or what is meant by a unity among
222 BOLTON :
a number of auditory impressions. What relation must
the impressions bear to one another that they may be
grouped together or grasped by a single act of apperception!
How may a number of impressions become the object of a single
state of consciousness? With Plato, we ask how the many
become the one, or with Kant, how the mind makes a unity
out of a manifold. Upon the basis of this study, we can hope
to answer the question with regard to auditory impressions
in a sequence. The question has already been answered for
simultaneous sounds — musical tones — in the laws of harmony .
The general principle as laid down in the treatment of poetry
was that by coordinating and subordinating the elements to
one another, unities were effected among them. The same
principle holds good here.
From the nature of the apparatus, only changes in the
intensity of the clicks could be effected. For this reason the
subordinations and coordinations among the sounds must be
accomplished through different intensities. Two methods for
determining the relations of the sounds in a group with
respect to their intensities were employed. By the first each
subject was asked how he effected a grouping in a series of
sounds which were of uniform intensity, and, if by accent,
what sounds in the groups were accented. By the second
method, the subject was given a series of sets of sounds of
different intensities, which recurred always in the same order,
and he was asked to point out where the series was grouped
— the position of the strong and weak sounds in the group.
In this way we were able to determine what was the most
natural order in which the different intensities occurred in
the group.
By the first method, it was determined that :
The first sound in the 2-group was accented. It was possi-
ble by objective suggestion of tapping, or counting, or by
voluntary effort, to accent the last sound, but no subject
would agree that this was the natural accent.
The first sound in the 3-group was strongly accented and
the second slightly. Occasionally a subject found it easier
to accent the second more strongly than the first, but
this did not seem to be the natural way of accenting the
group. It was possible by voluntary effort, or objective sug-
gestion, to change the position of the accent. Very few sub-
jects found it easy to group by three, and it usually required
a strong suggestion to start the group.
The 4- group was very generally accented upon the first and
third sounds ; the first was stronger than the third. There
was, however, some difference of opinion. Several subjects
found it easy and natural to accent the second and fourth, and
RHYTHM. 223
subject 15 was more inclined to this form of accent than to
the other. Sometimes there was only a single click accented,
and this was very generally the first. The accents could be
changed voluntarily. The reader is referred to the records of
subjects 1, 2, 6, 7, 9, 10, 11, 13, 15, 16, 17, 20.
Most subjects preferred a grouping by four to one by three.
When the attempt was made to suggest a 3-group by counting
three, they felt an over- powering tendency to count one or
three a second time. Thus: 1, 2, 3, 1—1, 2, 3, 1—1, 2, 3, 1
—or 1, 2, 3, 3 — 1, 2, 3, 3 — 1, 2, 3, 3. The former was the
more common. Subjects 1, 4, 7, 11, 12, 13, 15, 17 and 20
mentioned this phenomenon.
The 5-group was very difficult to suggest and maintain.
Most subjects declared their inability to get such a grouping.
Subject 11 said that an extra click would attach itself to the
group and < 'pull it over' ? to a 6-group. Subject 1 7 mentioned a
similar phenomenon. In counting a 5-group, it was found
easy to emphasize the first and third or the first and the
fourth. This gave to the 5-group the appearance of being
compounded of a 2-group and a 3-group. Subjects 10, 11, 12,
15, 16 and 17 make observations on their attempts to suggest a
5-group.
The 6 and 8-groups were generally compounded of smaller
groups of two, three or four. The 6-group was composed of
two 3-groups or sometimes three 2-groups. The first group
in the 6-group was more emphatic or was accented. The
8-group was composed of two 4-groups or sometimes four
2-groups. Subject 13 thought that the intensities of the
sounds in the 8-group decreased from the beginning to the
end.
Higher grouping of these groups was possible to some
extent. The most common form was to group by two. This
was spoken of as the pendulum- swing movement. In this case,
the first group was always accented. Subjects 1, 10, 12, 13,
15 and 16 make observations upon their attempts to group 2
and 3-groups. The record of subject 16 is especially impor-
tant. Several were unable to group 4-groups beyond two, on
account of their inability to keep the accents clear. In general
all subjects made a kind of interval between the groups. In 6
and 8-groups, which were compounded, a short interval fol-
lowed each smaller group and a longer interval followed the
whole group of six or eight.
Various methods of suggesting a grouping were employed.
The most frequent method was by counting or beating time
with the fingers. Subject 17 says: " Subjective counting is
most effective, or this assisted by respiratory stresses and
probably other muscular movements. " The associations
224 BOLTON :
which the sound brought up, very frequently suggested a form
of grouping. The clock (various kinds), pendulum, locomo-
tive, conical pendulum and revolving wheel, making a certain
number of sounds during a revolution, are most frequently
mentioned as influencing the form of grouping. The operator
frequently directed the attention of the subjects to respiration,
or asked them to feel the pulse. Most of the subjects incline
to the view that respiration accommodated itself to the form
of grouping that was found most natural with the rate to
which they were listening. Inhalation and exhalation each
lasted during the time of a 4 -group. In this way a kind of
higher grouping was accomplished, for the clicks heard dur-
ing inspiration were more intense. When the rate was slow
both inspiration and expiration were accommodated to the
time of one click.
With fast rates, the pulse acted as a suggestion. All the
clicks falling between two heart- beats were grouped together,
the click coming nearest in time to the heart-beat being
accented. Subjects 2, 4, 9, 10, 11, 13, 15, 17 and 25 make
observations upon this subject.
When the subjects were allowed to hear the sound of the
chronograph, which was distinctly rhythmical, no other
grouping was possible. The reader is referred to the records
of subjects 2, 3, 9, 11, 16, 19 and 23.
This general conclusion seems to be warranted : In the
presence of any fixed rhythm within limits, or of objective
suggestion, the series was grouped according to the sugges-
tion, and it was found difficult, if not impossible, to suggest
any other grouping. The grouping would follow the stronger
suggestion.
Certain rates were more favorable than others for voluntary
changes of the forms of grouping. Subjects 4, 6, 13 and 15
mention these rates respectively as especially favorable for
voluntary changes : .323 sec, .353 sec. and .268 sec. Sub-
jects 1, 2 and 16 thought that the grouping changed easily
when they were fatigued. When a very weak accent was
introduced every third in the series, subjects 4 and 13 did
not detect the accent, but grouped the series by three, and
were unable to suggest any other form of grouping ; but they
could not tell why the series grouped this way. Subjects 6,
9, 11, 12, 13, 15, 16 and 17 make observations upon easy
changes of the form of grouping.
The second method of determining the nature of rhythmical
groups was to give the subject a series which was composed
of a regularly recurrent set of sounds of different intensities.
Sets of two, three and four different intensities in groups of
two, three, four and five were studied. Very few observa-
BHYTHM. 225
tions were made upon 5-groups. To make a graphic repre-
sentation of such series of sounds, let A, B, C and D represent
the four intensities of sound, A the strongest, and D the
weakest. By using only two intensities (A B) it is possible
to form the following series of sounds :
2-groups. ABABABAB (1)
q c«.™™ / ABBABBABB (2)
3-groups. JAABAABAAB (3)
fABBBABBB (4)
4-groups. 4AABBAABB (5)
(AAABAAAB (6)
Of 5-groups, these only were tried
ABBBBABBBB (7)
ABABBABABB (8)
AABABAABAB (9)
The question was to determine where the mind most
naturally made the division into rhythmical groups. The
first series might divide in two ways, thus : A B — A B,
or B A— B A. The second in three : A B B— A B B, or
B A B— B A B, or B B A— B B A. Details regarding the
others are unnecessary. Of series composed of three intensi-
ties, the following out of all the possible forms were thought
to be characteristic, and were tried :
q ,™™„ / ABO ABC ABO
3-groups. JAOBAOBAOB
4-groups. AOBC AOBO ACBC
Of series of four intensities, the following out of the many
possible forms were tried :
a ,™vn™ / ABOD ABCD
4-groups. | ADOB ADOB
In the following table are given the rhythmical groups which
each subject made of the series upon which he was tried. At
the top of the table, in each column, are letters which indicate
the order in which the different intensities recurred in the
various series. The number of each subject is given in the
left-hand column. If a subject has given a stronger intensity
to a click than it actually possessed (said that a B intensity
was equal to an A, or a C to a B), it is printed in full-faced
type. Where a subject has remarked upon a longer interval,
either following or preceding the strongest sound, this is
indicated by placing a dash either before or after the strong
sound :
W
PQ
PQ
PQ
<
«*
<
<
PQ
PQ
PQ
PQ
o
<3
<3
^
<3
*
1
<J
<
<J
<l
PQ
PQ
PQ
PQ
PQ PQ
PQ
PQ
PQ
1
PQ
PQ
PQ
PQ
^ PQ
PQ
PQ
PQ
£
I
p
«!
<
<
<j
PQ <
<J
<
<J
0*
PQ
PQ
PQ
PQ
PQ PQ
PQ
PQ
PQ
§
o
15
<j
«!
^
<
«! <
<
<J
<J
PQ
PQ PQ
PQ
PQ
PQ
PQ
PQ
PQ
PQ
PQ PQ
PQ
PQ
«!
PQ
PQ
PQ
PQ
PQ PQ
PQ
PQ
PQ
PQ
PQ
PQ
PQ
PQ PQ
PQ
PQ
PQ
PQ
PQ
PQ
<4
< <
<
<
PQ
<
<
<
PQ
<
O 6^
Oft pi
i
ft
ft
ft
ft
s
O
PQ
^ ft<J
GH O
<J
<J
Jl
<
•I
1
s
a
(4
ft
O
PQ <tjpq
^PQ ft
PQ
PQ
PQ
PQ
<4
ft
O pqOPQO <tj
O
O
O
O
ft
O
ft
ft ft i
i i
ft ft
i
ft
i
ft
ft
ft
I
O
PQ
O
O^D q
o o
O
O
O
O
R
PQ
<
PQ
PQ«PQ pq
PQ PQ
PQ
PQ
PQ
PQ
6
<
ft
<
<J ^J 4
< <
<
<
<«i
<
03
O
O
DO O o
i
O
O
6o
Oq
O
09
2*53
PQ
PQ
<JpQ PQ pq
PQ
PQ
PQ^
«<1
PQ
61
O
O
DO O o
O
O
DO
°o
O
S
5
<
i
PQ<< < «s*
<
<
<t{PQ
<&
<
PQ
<
PQ
PQ pi
i i
PQ PQ
n<
PQ
PQ
PQ
i
p
<J
<
<8
<! <
<S <j
<<
<:
-tS
^J
!
<4
<:
<
« ^
< <
<<j
<«J
<J
<J
I
<«J
PQ
<
« ^
< <
^PQ
^
<
<
PQ
PQ
PQ
« M
pi
i
PQ
PQ
PQ
PQ
1
PQ
PQ
PQ
^ «
PQ
PQ
PQ
M
PQ
I
<
<
<4
^ <
<
<
<
<^
^
1
<
<
<
PQ ^
<
<
<
^J
<^
PQ
PQ
«
PQ ^
p^ A
i
PQ
PQ
««
PQ
PQ
PQ
M
PQ pq
PQ PQ
PQ
PQ
PQ^
PQ
PQ
<
PQ
M PQ
PQ PQ
PQ
PQ
pqPQ
PQ
<
PQ
<4
<J <
< <
<
<
<JPQ
<
i
PQ
O
O O <J
6 6&
\ o
i i
OPQ
OPQ
O
i
o
O
i
i
O
<
«} «! PQ
<< <o <
«*?
<1o
1
<
<
<
i
<
PQ
PQ W O
PQ PQ<
\ «
PQ<1
W<J
PQ
PQ
PQ
O
O
O O O
6 6
o
6
O
mo
i
O
O
I
PQ
PQ
PQ PQ PQ
PQ PQ
PQ
PQ
PQ
<ii«
PQ
PQ
H
<4
<
«! <4 oj
< <
<
4l
<
oJi
<
<l
i
1
*
PQ
A Pi
PQ
PQ
p*
1
CO
1<
< <
<
<
PQ
PQ PQ
PQ PQ PQ
Pi Pi
PQ
«
PQPQ
n
PQ
PQ
o
PQ
PQ M
PQ PQ PQ
PQ PQ
PQ
PQ
PQ<1
PQ
PQ
PQ
<4
< <
<4 <4 <
< <
<
^
<&
Jl
<J
<
sg
iH CM
CO
rt* iO CD t<- O*
O iH
dCO ^
i©
t*
CO
00
iH iH
y-*T* iH
iH
iH
W
w
RHYTHM. 227
At the bottom of each column is given the form of group
which is generally made when the series is formed in the way
indicated by the letters at the top of the column. Two factors
seem to operate in determining where the series shall be
divided into groups. The group must begin either with
a very intense sound or close with a very weak one. The
subject strives either to put all the strong sounds as near the
beginning as possible, or all the weak ones as near the close
as possible. There are three cases where these principles
are brought into strong conflict. The first is where the series
is composed of three intensities, in the order of A C B or C
B A. Either the strongest cannot come first, or the weakest
last. The weakest generally comes last and the strongest
second; the second being the position of the secondary
accent in a musical rhythm. The other form which is com-
mon and sometimes preferred, places the strongest first and
the weakest second. This does violence to both the weakest
and middle intensities, by placing an unaccented sound in a
position that requires an accent, and an accented sound in a
place where an accent does not occur. A second case is where
the series is composed of four different intensities in the order
of A D C B or D C B A. The strife would be greater here
than in the first case, if it were not for the fact that the third
position may frequently receive the sound of the greatest
intensity. The most common form is to place the weakest
sound last, and the strongest in the third place. This of
course leaves the first place occupied by an unaccented
sound, though this sound is stronger than the last. Subject
9 makes the strength of the first equal to that of the second,
and thus harmonizes the group somewhat with the 4-group,
which is formed of uniform sounds. The third case is where
the series is formed of a sequence of 4-groups of three differ-
ent intensities, thus : A C B C. There is really no strife
here, except that the strongest sound often appears in the
third place. There are some irregularities, but none sufficient
to require special notice. The general principle just laid
down is well illustrated in the last two forms of the 5-group,
when composed of two intensities of sound.
When two or more strong sounds, standing together, are
followed by a weaker sound, the sound which is followed
immediately by a weaker one appears stronger in contrast
with the following weaker sound than the preceding, which is
actually of equal strength. This will be observed in every
case with 4-groups composed thus : A A A B. And in
one case, with the 3-group composed thus : A A B. The
third sound in the 4-group and the second sound in the
3-group appear to be stronger than the preceding sound. A
228 BOLTON :
further fact to which attention is called, is the long interval
which appears between the groups. The pause seemed to be
due to the fact that a long interval generally preceded the
accented sound. At the same time some subjects, especially
10 and 15, make a short interval after the strongest sound.
To most subjects, the strongest sound seemed longer than
the rest. With some this was more apparent than with
others. Subjects 1, 3, 4, 10, 11, 13, 14, 15, 24, 27 and 28
either confound the accented click with a longer interval, or
make the louder click seem longer than the others. Subject
11 speaks of the strongest sound spreading itself over the
rest. It is possible by voluntary effort to avoid the illusion
of a longer interval, either preceding or following the accented
sound, but ordinarily it was very clear. When the strength
of all the sounds in the series was increased, the rate seemed
slower. Subject 27, especially, makes this observation.
When these accented 3 and 4- groups were given at a fast rate
(.134 sec), the separate clicks seemed to fuse into a single
impression, which grouped generally, by four.
The different intensities of sound bore no special relation
to one another ; the strongest was clearly discernible from the
second, and the second from the third, and so on. When in
a series of impressions of the third (C) intensity, the strongest
(A) was introduced every fifth, it seemed to appear as an
extraneous sound which would not group with the others.
(See records of subjects 9 and 11). It appeared from several
other records also that sounds differing greatly in intensity
would not easily group together. When a very weak accent
was placed upon every third sound, subjects 4, 13 and others
did not discover the accent ; they expressed their inability to
group the series in any other way, but could not understand
the reason. Subjects 13, 16 and 17 expressed the opinion
that strong accents were disagreeable ; they preferred their
own accents to real accents of any strength. Eeal accents did
not seem to form so harmonious a group as did the accents
which the subjects put in themselves.
As a further investigation into the nature of rhythmical
groups, especially with reference to poetical rhythms, it was
proposed to employ sounds of which the length or endurance
might be varied. The click of the telephone is almost
instantaneous. The disk probably makes a very few vibra-
tions. We are indebted to Dr. B. C. Sanford for devising
and constructing an apparatus, which served the purpose
admirably in some ways. The principle involved in this
apparatus was simply interrupting the sound of an electric
tuning fork, which was placed before the opening to one of
Helmholtz's resonators. When a card is placed over the
RHYTHM. 229
opening into the resonator, which is near a tuning fork of the
same pitch, the sound of the fork is rendered almost
inaudible. Eegular interruptions result in a series of uniform
sounds and silences.
Figure IV. 1
For this experiment were required an electric tuning fork
and a set of disks with notches cut in the circumference.
The resonater rested horizontally, supported by a stative
near the edge of a small table, upon which the tuning fork
was placed. The fork and the resonator were placed at the
same height, with just enough space (about half an inch)
between the end of the fork and the opening of the reso-
nator to allow a pasteboard disk to pass without interference.
This disk was about twenty inches in diameter, and placed
at just the proper height to cover up the opening into the
resonator. Notches were cut in the circumference of the
disk in such a way that when it was revolved the opening
into the resonator was now closed and now open. With a
regular revolution of the disk, and with notches of an equal
number of degrees, and equal spaces, a series of sounds, uni-
form in length, pitch and intensity, would be produced. To get
sounds of different lengths, some notches were made to cover a
greater number of degrees of the circumference than others.
ir This represents the apparatus as it has since been perfected by
Dr. Sanford.
230 BOLTON :
The number of degrees in some cases was twice that of others.
By cutting the notches upon the arc of a circle, and at just a
sufficient depth to cover a part of the opening of the reso-
nator, it was possible to decrease the strength of the resonance,
and thus get a sound of less intensity. This gave the effect
of an accent upon certain sounds by weakening others.
The following are the forms of disks that were thought to
be characteristic. Just enough space was left between the
notches to cause a silence in the sound of the fork. The
spaces were always of the same number of degrees in a given
disk. Notches were cut in the circumferences of the disks
as follows :
1. Two notches, each of 150 degrees. One accented.
2. One notch, 200 degrees, and one, 100 degrees.
3. One notch, 200 degrees — accented — and one, 100 de-
grees.
4. Three notches, each of 100 degrees.
5. Three notches, each of 66 degrees. The spaces of 40
degrees.
6. Three notches, each of 100 degrees. One accented.
7. One notch, 120 degrees, and two, each of 60 degrees.
8. One notch, 120 degrees — accented — and two, each of 60.
9. Four notches, each of 60 degrees. Strong accent upon
the first, and weak upon the third.
The question was to determine where the rhythmical groups
began, with the long or the short sounds. As the pause
between the successive sounds was the same length, it is a
matter of interest to determine what effect the rhythmical
group had upon the pause coming after the sound with which
the group closed.
The results of the experiments with the different series of
sounds produced by the disks, as described above, are given
in the following table. A long sound is indicated by a capital
letter, and a short one by a small letter. In disk 9 the ac-
cented sounds, which were of different intensities, are repre-
sented, the stronger by a bold faced A, and the weaker by a
bold-faced B. An accented sound is given in full-faced type.
In turning the disk, the operator sought to keep a uniform
speed at a moderate rate — about one turn to the second :
RHYTHM.
231
2-Qroups.
3-Groups.
4-Groups.
1
A A
2
A a
3
Aa
4
AAA
5
AAA
6
AAA
7
A a a
8
A a a
9
A A B A
1
3
4
7
10
11
15
17
18
27
28
A A A A
A A
AA A A
A A
A A
A A
a A
A a A a
a A
a A
A a
a A
a A a A
a A
a A
a A
a A-
a A
a A
A a
a A
a A
a A
a A
a A
AAA
A A A A
A A A A
AAA
AAA
A A A A
A A A A
A A A A
A A A A
AAA
AAA
AAA
A A A A
No group
AAA
A A A A
4- group
or
3- group
AAA
AAA
AAA
AAA
AAA
A A -A
AAA
AAA
A A A-
AAAA
AAA
AAA
a a A
A a a
a a A
a a A
a a A
a a A
A a a
a a A
a a A
a a A
a a A
a a A
a a A
a a A
A a a
a a A
a a A
a a A
A A B A
B A A A
B A A A
A A B A
A B A A
A A B A
Groups
prefer'd.
A A
a A
or
a A
AAA
a a A
a a A
A A B A
Several facts are to be observed in this table. First, a
series of sounds of uniform length and intensity may be
grouped by two, three or four. With disk No. 4, while the
most common form of grouping was by three or four, by
turning very slowly it was possible to group by two, or by
turning faster to group by six or eight. With No. 1 it was
easy to group by two or four by turning slower or faster.
When disk No. 9 was turned at a slow rate, the sounds were
grouped by two, at a faster rate by four, and at a still faster
rate the 4-groups were grouped by two or by four.
Second, a more intense sound occurring regularly, imposes
a grouping according to the number of sounds between the
accents. The accented sound comes first in the 2 and 3-groups,
and in the 4-group the first and third receive accents. The
first is more strongly accented than the third.
232 BOLTON :
Third, a longer sound occurring regularly in the series,
imposes a grouping according to the number of sounds
between the longer ones. The long sound, as a rule, is
the last in the group, and is frequently accented. It was
possible for most subjects to change the place of the long
sound to the first of the group, but with the exception of
subject 11, it was difficult to keep it at the beginning of the
group. Most subjects remarked upon the long interval or
pause which seemed to follow the long sound, and for this
reason it was found difficult to make the close of the group
come at any other place. When the attempt was made to
begin the group with the long sound, the preceding group
would not seem to separate from the following; the two
would run together and become indistinguishable. In the
telephone experiments, when a subject attempted to suggest
a 3-group, which was accented upon the third by counting
one, two, three, emphasizing three, it required the closest
attention to make the group close with three, for the emphatic
three would begin the group thus : Three, one, two, etc.
Although it was impossible to control the rate, faster rates
than common caused these groups to group by two or four.
The accented long sound frequently appeared more pro-
longed than the unaccented sound of the same length ; the
accent had the effect both to increase the length of sound and
of the interval which followed.
When the short sound in disk No. 2 and the last short
sound in disk No. 7 were accented, the accented sound always
came first and the long sound last. It was more difficult
with this arrangement to place the long sound first and the
accented last, than before.
The results of this experiment confirm in part the results
of previous experiments concerning the nature of rhythmical
groups. First, the accented sounds occupy the first place in
the group. Second, the weaker accent comes upon the third
sound in the 4-group. Fast rates with accented groups
caused them to fall into higher groups, first of two, and then
of three or four.
We come now to the consideration of the nature of the
rhythmical group. The general principle is this : In a
series of auditory impressions, any regularly recurrent im-
pression which is different from the rest, subordinates the
other impressions to it in such a way that they fall together
in groups. If the recurrent difference is one of intensity, the
strongest impression comes first in the group and the weaker
ones after. If the recurrent difference is one of duration, the
longest impression comes last. These rules of course hold
good only within the limits spoken of above. When the im-
pressions are uniform in length and intensity, the mind
EHYTHM. 233
enforces a grouping by giving fictitious values to the impres-
sions, generally with respect to intensity, but sometimes with
respect to duration. At the rate .795 sec, the mind intensi-
fies every other sound, so that the series is grouped by two.
The second sound in the group is subordinated to the first.
At the rate of .460 sec, the mind finds it easy to group a
series of auditory impressions by three, by intensifying the
first greatly and the second slightly, so that the second is
subordinated to the first and the third to the second. More
than three degrees of intensity do not appear together in the
order of their intensities in a series. In grouping by four,
which takes place generally at the rate .307 sec, the mind
accents the first strongly and the third slightly. The second
and fourth impressions are generally of the same intensity.
If there is any difference in intensity, the second is stronger
than the fourth, but it is always less than the third or the
first. It would appear from this that the 4-group is com-
pounded of two 2-groups, or it may perhaps arise, as Haupt-
mann says in his "Natur der Harmonik und Bythmik," from
a combination of two 3-groups. However this may be, the
4-group does appear as a harmonious and organic unity in
itself. Given, then, a series of impressions which is made
up of three or four intensities recurring as a sequence of
fours, the mind divides the series into rhythmical groups,
whatever may be the arrangement of the intensities in the
sequence of four, so that the impressions are subordinated to
one another as nearly as possible from the beginning to the
end. The effort is always made to subordinate the last im-
pressions to the first. The same holds good for series which
are made up of sequences of two or three. In a sequence of
twos, only two impressions can recur ; the stronger is always
first in the group. In a sequence of threes, the groups may
contain two or three different intensities, but the mind always
divides the series in such a way that either the strongest
comes first or the weakest last.
When the series is composed of impressions different in dura-
tion (the longer impression twice the length of the shorter),
recurring in a sequence of twos, the mind groups the series by
two, placing the longer impression last, and at the same
time gives to it frequently a greater intensity. When the
series is composed of a sequence of threes, one long and two
short, the mind groups the series by three, placing the longer
sound last, and at the same time giving to it also frequently
a greater intensity. The order of subordination is here
reversed. The more important element in the group comes
last. For this fact we can offer no explanation upon purely
psychological grounds. The fact, however, is interesting for
its connection with poetry. Although, as we have seen,
234 BOLTON :
English poetry in its early history contained feet accented
upon the first syllable, the most common foot in modern
poetry is accented upon the last syllable. What formerly
was the beginning of the foot is now the end. In the experi-
mental study wibh long and short sounds — these correspond
to syllables — ali the subjects found great difficulty in not mak-
ing a pause after the long sound, which compelled them to
begin the group with the short sound. It was impossible to
avoid this pause or to make another after the short sound
•equal to it, although the interval in every case was the same.
Upon this basis and other facts mentioned above, we are able
to base our answer to the question whether there is a foot-
division in English poetry. Although the long and short
syllables do not stand in the absolute relation of two to one,
yet the syllables do differ in length and in intensity of accent,
and for that reason they tend to fall together in groups. The
accented syllables, like the accented sound, will seem to be
longer than the unaccented, and in uttering them the speaker
will prolong them still farther. Series of syllables, then,
which are arranged with reference to the regular recurrence
of the accented syllables will fall into groups, and since the
accented syllables are longer than the unaccented, a pause
will be felt after the long syllable. To use the Latin termi-
nology, the most natural foot must be either iambic or
anapaestic. This, however, seems to be due largely to
modern ways of utterance. In order for a word to be intelli-
gible, it must be distinctly and carefully enunciated. In the
early history of poetry, it was always recited in highly
emotional states ; words were not articulated, they were
houted. The line of poetry was little more than a series
of strong and weak sounds, which, we can argue upon
the basis of our experiments, would be grouped with the
strongest first and the weakest last. In the change from the
merely emotional shout to articulated utterance, the character
of the foot changed from one which was accented upon the
first to one which was accented upon the last.
We come now to the subject of muscular movements and
their relation to rhythms. Most subjects felt themselves
impelled by an irresistible force to make muscular move-
ments of some sort accompanying the rhythms. If they
attempted to restrain these movements in one muscle, they
were very likely to appear somewhere else. Wundt 1 says
that the intensive clang change has its nearest pattern in the
sensation of motion. A corresponding rhythmical series of
motions associates itself in dancing, marching and beating
^hysiologische Psychologie, Vol. II. p. 73.
RHYTHM. 235
time, with almost irresistible force to the changes of strength
in the clang.
The most common forms of muscular movement were beat-
ing time with the foot, nodding the head, or swaying the body.
Subjects 3, 10 and 17 accompanied the rhythmical grouping
by muscular contraction of the diaphragm and chest, and it
was exceedingly difficult to restrain them. Other subjects
counted inaudibly or made the proper muscular adjustments
for counting. Slight or nascent muscular contractions were
felt in the root of the tongue or larynx. Most subjects were
unconscious of their muscular movements until their attention
was called to them, and subject 15 never became conscious of
the rhythmical contractions in the eyelids. When he was
asked to restrain all muscular movements, he found great
difficulty in maintaining the rhythmical grouping. This fact
was remarked upon by other subjects also. The reader is
referred to the records of subjects 2, 3, 7, 9, 10, 11, 12, 13
and 15.
Of the same nature as muscular movements, are the asso-
ciations of various objects. Most subjects visualized the
pendulum and clocks, large and small. Several referred to
the conical pendulum, striking three or four times in a swing,
and others to revolving wheels. Subject 14 visualized a
series of dots, and subject 11 at one time an undulating line,
and at another an ellipse with four dots placed upon either
side. Subject 15 made a color association.
The question we have to decide upon is, are these muscular
movements and associations the result or the conditions of
the rhythmical grouping f With Eibot we accept without
hesitation the latter.
Eibot states this principle, " Every intellectual state is
accompanied by physical manifestations. 1 Thought is not —
as many, from tradition, still admit — an event taking place
in a purely supersensual, ethereal, inaccessible world. We
shall repeat with Setchenoff, ' No thought without expression, 7
that is, thought is a word or an act in a nascent state, that is
to say, a commencement of muscular activity. 7 ' Each im-
pression as it enters into consciousness tends to find expres-
sion in a muscular movement, but the intensive changes in
the series of impressions produce corresponding changes in
the intensity of the sensations, which must find expression in
different degrees of muscular activity. In order to express
these different degrees of sensation, the muscular movements
must rise above the merely nascent state in which they
ordinarily occur, and manifest themselves in visible muscular
a The Psychology of Attention.
236 BOLTON :
movements. The tendency for sensation to find expression
in visible muscular movements is stronger with children and
primitive peoples than it is with highly civilized and
especially* well-trained persons. With the latter class, mus-
cular movements accompanying attention do not so easily
rise above the nascent state.
Exact coordinations of sounds with respect to intensity are
difficult, for the reason that great degrees of difference must be
allowed, that two sounds may be discriminated. This is proved
by the fact that higher groupings of 4-groups are difficult, for the
reason that the differences in the accents cannot be kept clear.
Groups of six and eight are difficult because the different
degrees of intensity required cannot be discriminated. Pitch
changes are much more easily discriminated, and more exact
coordinations are possible. They find their expression in
different degrees of tension in the muscles of the larynx.
With fast rates the intensive changes recur more rapidly,
and hence call for more rapid muscular movements. On this
account the faster rates were found exhilarating and animating,
and the slower rates drowsy and soporific.
For the same reason, subject 12 found that a change from
a 3-group to a 4-group gave rise to a feeling of a slower pace.
Within certain limits the mind can easily accommodate itself
to changes of rate. A rate which seemed unpleasantly slow
or fast at first, became in time pleasant. If the rate is slow,
the grouping which is first suggested is accompanied by a
feeling of suspense — subject 11 said the group broke off with
a " dead end " — but if it is fast there is a straining after a
longer group, or perhaps a hurried, animating feeling which
becomes monotonous. If a subject maintained a 2-group, for
instance, with a rate which was naturally too fast for group-
ing by two, it became exceedingly monotonous in a short time.
If the length of the group corresponds to the normal wave
of attention, the grouping gives rise to a feeling of satisfaction
and repose. There is probably not an absolute psychic con-
stant in attention which admits of no variations without feel-
ings of dissatisfaction, but within limits a constant is easily
established, which, if changed gradually, accommodates itself
to a longer or a shorter interval. A sudden change, however,
cannot take place without difficulty. For this reason, if the
grouping enforced by an irregular recurrence of an accented
sound change rapidly from one form of grouping to another,
it gives rise to an alternation of feelings of suspense and
straining which no one fails to perceive. The same phenom-
enon would arise if the temporal sequence of the impressions
were irregular. Either it would be necessary to group now
by three and now by four, or by two, that the interval between
RHYTHM. 237
the successive accents should be the same, or there would be
an alternation of feelings of suspense and of straining to
maintain a grouping by three or any other number. When the
rate was changing rapidly, as it did just after the chronoscope
was started (it required ordinarily about two minutes for the
chronoscope to attain its full speed), subject 4, especially,
and others remarked upon the diagreeable effect. The
accommodation to any form of grouping within certain limits is
easy, providing there is a perfect regularity in the sequence.
The accents must recur at regular intervals, and the number
of intermediate impressions remain the same, or there is no
feeling of rhythm. When a slow rate was succeeded by a
faster one, it gave rise generally to a disagreeable effect ; but
in time the subject could accommodate himself to it. Subjects
4, 9 and 15 make observations upon this point. Subjects 2
and 5 were greatly puzzled over a 5-group which was accented
on the first and third sounds. They attempted to group by
two and by three alternately, which gave rise to a very disa-
greeable feeling:. When, however, they grasped the regular
sequence of five, the disagreeable feeling passed away.
When a longer interval was introduced into the series, the
impressions coming between the long intervals fell together
into a group, but they did not form an organic unity. There
was no pleasure in such a rhythm. Something seemed to be
looked for in this longer interval which was wanting. When
the rate was made very fast, the impressions between the
long interval seemed to fuse together into a single impression
and then to group by two or four.
This general principle may be stated : The conception of
a rhythm demands a perfectly regular sequence of im-
pressions within the limits of about 1.0 sec. and 0.1 sec.
A member of the sequence may contain one or more simple
impressions. If there are a number of impressions, they
may stand in any order of arrangement, or even in a state
of confusion, but each member of the sequence must be
exactly the same in the arrangement of its elements.
The application of this principle to poetry demands that
the accents in a line shall recur at regular intervals ; it
requires also that the successive feet in a line shall be of pre-
cisely the same character. The introduction of a 3-syllable
foot into an iambic verse is allowable on this condition only,
that the 3-syllable foot can be read in the same time of the
two, so that there shall be no disturbance in the temporal
sequence of the accents. This foot affects the rhythm in so
far only as it changes the character of one member of the
sequence. This is an actual disturbance to the rhythm, but
it is allowable for the purpose of emphasis. The frequent
238 BOLTON :
use of such a foot would be fatal. Poe's principle that the
regular foot must continue long enough in the line, and be
sufficiently prominent in the verse to thoroughly establish
itself, is perfectly valid. In a musical rhythm, however, the
measures may vary with certain restrictions in the arrange-
ments of their elements. But it is just this variation which
constitutes the melody to a certain extent. The rhythm is
varied for purposes of melody, but it is, nevertheless, a dis-
turbance to the rhythmical flow in so far that it changes the
measure. The melody is a new and higher unifying agency,
which corresponds in a way to the use of rhymes in poetry.
The temporal sequence of the accents is always preserved.
It remains now to make my acknowledgment to those who
have assisted in the work.
To President GL Stanley Hall I am indebted not only for
the subject itself, but for a large amount of material which
he had already collected upon it ; also for suggestions as
regards the direction of the experiment and references to
literature.
To Dr. E. C. Sanford, the director of the laboratory, is due
much of the credit for the success of the work. But for his
skill in devising and constructing apparatus, the work could
not have been carried on. His suggestions as regards
methods for making the experiment were no less valuable
than his assistance in devising apparatus. To all others who
so generously gave up their time to sit through long and
tedious experiments, I acknowledge my indebtedness. Space
forbids me making special reference to each one.
Clark University,
Worcester, Mass.
August, 1893.
Live Music
Archive
Librivox
Free Audio
Metropolitan Museum
Cleveland
Museum of Art
Internet
Arcade
Console Living Room
Open Library
American
Libraries
TV News
Understanding
9/11